Datasets:

---

CodeGoat24

/

LiFT-HRA

like 0

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a futuristic setting, a police officer, clad in a high-tech suit with neon accents, runs across a rugged asteroid surface illuminated by the distant stars. The officer, equipped with a helmet and breathing apparatus, sprints past mining equipment and industrial structures, indicating an asteroid mining operation. With each stride, the officer's suit glows under the harsh lighting of the asteroid's environment, highlighting the action and determination in this dynamic scene set against the backdrop of space exploration and law enforcement.]" }, { "from": "gpt", "value": "Bad. The officer's running in the video is very unsmooth, and the movements are uncoordinated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In the heart of Tokyo, under the neon lights that bathe the bustling streets, a lone coyote wanders. Its golden eyes reflect the urban glow as it moves through the crowded alleyways. It pauses, the camera capturing a close-up of its face, showing the wild creature in an unexpected environment. The coyote then proceeds to wipe its face with a paw, a moment of natural behavior in stark contrast to the man-made surroundings. Behind it, the signs in Japanese and English, the busy storefronts, and the occasional passerby underscore the unlikely scene of a wild animal navigating the concrete jungle.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged janitor, balding with glasses and a neatly trimmed beard, wearing a dark green collared shirt, focuses intently on fixing a leaky faucet in a public restroom. The faucet is positioned above a stainless steel sink with a mirror in the background. The janitor kneels on a yellow floor mat, using a wrench from his well-stocked tool belt to tighten the faucet's handle. The scene is illuminated by overhead fluorescent lights, casting a warm glow on the janitor's determined face as he works diligently to stop the dripping water.]" }, { "from": "gpt", "value": "Bad. The actions are not continuous, there are sudden changes." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A cheerful musician, donning a yellow shirt and dark glasses, vacuums the floor of a bright room filled with musical instruments. The scene opens with the musician, wearing a yellow t-shirt, dark glasses, and a smile, vacuuming the carpeted floor. The room is illuminated by sunlight streaming through a large window, casting a warm glow on the various instruments scattered about, including guitars, drums, and keyboards. As the musician moves the vacuum cleaner across the floor, they pause to adjust a guitar strap hanging from a nearby stand, revealing a passion for music even during mundane household tasks. The final shot captures the musician, still smiling, vacuuming near a drum set, highlighting the harmonious blend of daily chores and artistic pursuits within the vibrant space.]" }, { "from": "gpt", "value": "Good. Basically no problem." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist, wearing a brown beret and glasses, sits in a dimly lit studio filled with art supplies and paintings, engrossed in reading a book about art history. The camera captures the artist's hands turning the pages, revealing colorful illustrations of famous artworks. In another shot, the artist is seen from behind, sitting in front of a large canvas with paintbrushes in hand, contemplating the next stroke while a book on art techniques lies open nearby. Throughout the video, the artist's passion for both creating and studying art is evident in the careful attention given to both the book and the canvas.]" }, { "from": "gpt", "value": "Normal. The artist's movements in the video are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a unique fusion of scenes, a waiter in a crisp white shirt and black trousers serves a golden, perfectly roasted chicken on a rustic wooden table set amidst the lush greenery of a temperate rainforest. Moments later, the waiter, now donning a vibrant, traditional costume, gracefully dances among the towering trees, his movements harmonious with the serene environment. The chicken, now artfully presented on a platter, sits center stage as the waiter performs a captivating dance, his attire reflecting the rich cultural heritage associated with temperate rainforests, all under the canopy of majestic, moss-covered trees.]" }, { "from": "gpt", "value": "Bad. The character's movements in the video are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A uniformed police officer, donning a navy blue cap and sunglasses, runs along the sandy shore of a bustling beach resort, where vibrant umbrellas and lively beachgoers fill the background. His attire includes a navy blue shirt with badge, navy blue trousers, and black shoes. The setting is a crowded beach during daytime, with the officer's movement capturing the essence of duty amidst leisure. He sprints past colorful umbrellas and animated vacationers, highlighting the contrast between his purposeful stride and the relaxed atmosphere of the resort.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A chef, dressed in a classic white uniform and black pants, walks through a lush, green forest carrying a live rabbit in his arms. The chef's attire is detailed, complete with a traditional white chef's jacket, black trousers, and a pristine white chef's hat. The rabbit, a live and lively creature, is held gently in the chef's arms as he strolls through the dense woodland. The forest setting is richly depicted, with tall trees, vibrant foliage, and dappled sunlight filtering through the leaves, creating a dramatic contrast with the chef's uniform and the rabbit's fur.]" }, { "from": "gpt", "value": "Bad. The chef's hand in the video is deformed, and the rabbit is generated abnormally, with its ears suddenly twisting and deforming in the middle frames." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A large, solitary rhinoceros is seen swimming in a natural body of water, surrounded by lush green vegetation. Its body is partially submerged, showing off its powerful swimming capabilities as it moves through the water. The rhino's head remains above the surface, its eyes focused ahead, and its mouth occasionally opening to breathe. The water gently laps at its thick skin, creating ripples that extend outward. In the background, tall trees and dense foliage provide a contrast to the open water where the rhinoceros swims.]" }, { "from": "gpt", "value": "Normal. The rhinoceros's mouth and ears in the video have generation issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A prehistoric mammoth, remarkably preserved and glowing with bioluminescence, orbits the Earth in a dreamlike sequence, passing over a city skyline with bright lights twinkling below. The scene transitions to the mammoth swimming gracefully underwater in a vast, clear ocean, its massive body moving effortlessly among schools of colorful fish. The camera then pulls back to reveal the mammoth floating in space once again, the blue planet Earth serving as a majestic backdrop, before returning to the serene underwater environment where it continues to swim peacefully, a surreal juxtaposition of ancient life and cosmic wonder.]" }, { "from": "gpt", "value": "Normal. Missing the scene of the mammoth swimming underwater." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A compassionate teacher, wearing a gray shirt and dark pants, sits cross-legged on a green yoga mat placed atop a lush lawn, gently caressing a small vole. She holds the tiny creature in her right hand, her face reflecting warmth and care, as she looks down at it lovingly. The vole, nestled in her palm, seems relaxed and comfortable, its presence offering a tender moment of connection between human and nature, under the shade of a large tree.]" }, { "from": "gpt", "value": "Bad. The character's face is distorted, hands are twisted, and feet are deformed." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A large hippopotamus, identifiable by its gray skin and rotund body, is seen walking through a shallow river, creating ripples in the water with each step. Its small ears, round eyes, and wide mouth are clearly visible as it moves slowly through the riverbed. The scene is set against a backdrop of lush vegetation and a clear blue sky, with the water reflecting the greenery and the hippo's movements. The hippopotamus continues its leisurely walk, leaving a trail of disturbed water behind it, all captured in high definition to showcase the details of this magnificent creature in its natural habitat.]" }, { "from": "gpt", "value": "Normal. It did not generate a clear blue sky." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A police officer, clad in a navy uniform with a badge and gun holster, leaps into the air in the middle of a vast, temperate grassland. The landscape is dominated by knee-high green grasses and a few scattered oak trees, with the horizon stretching far into the distance. The officer's feet leave the ground, capturing a moment of action and energy, as the rolling hills and clear blue sky provide a serene contrast to the dynamic scene. The officer's movements are sharp and decisive, embodying the spirit of law enforcement in the open, wild environment of the grasslands.]" }, { "from": "gpt", "value": "Bad. The character's limb proportions in the video have issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A rugged firefighter, donning a black helmet and protective gear, sits intently at a desk in a dimly lit station. His strong hands, adorned with a silver ring, deftly type on a laptop keyboard, while a yellow fire truck is visible through the window behind him. The scene conveys a sense of duty and focus as the firefighter's gaze is locked on the computer screen, illuminated by the soft glow of the monitor. A framed photo of a family rests on the bookshelf in the background, adding a personal touch to the otherwise utilitarian space.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dimly lit, futuristic room filled with advanced technology and screens displaying alien landscapes, a renowned director, dressed in a black turtleneck and jeans, sits in a swivel chair, intently observing a large screen. Suddenly, an otherworldly creature, resembling a cross between a mouse and an alien, emerges from the shadows. With a look of wonder and curiosity, the director gently extends a hand, initiating a tender caress of the creature's soft fur, highlighting a peaceful and unexpected interspecies connection. The scene is framed by high-definition visuals and enhanced by the ambient glow of the tech-filled environment, emphasizing the director's calm and respectful interaction with the unique being.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical scene, a friendly capybara wearing a stethoscope and a lab coat, akin to a doctor, energetically jumps in slow motion against a vibrant green meadow background, capturing the joy and playfulness of the moment. The capybara's enthusiastic leap is highlighted by its joyful facial expression and the picturesque setting, blending elements of fantasy and nature.]" }, { "from": "gpt", "value": "Bad. The capybara's legs disappear when it jumps, and then reappear." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist wearing a blue denim jacket and black cap, holding a paintbrush in his right hand, walks alongside a large elephant with intricate, colorful patterns painted on its body. Both the artist and the elephant stroll through a lush, green landscape filled with tall trees and vibrant vegetation. The elephant carries a canvas on its back, and the artist occasionally glances at it as they walk together in harmony. They are surrounded by the natural beauty of the environment, which complements the vivid artwork on the elephant's skin.]" }, { "from": "gpt", "value": "Bad. The elephant's walking movements in the video are uncoordinated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An engineer, donning safety glasses and a gray shirt with a vest, is depicted laughing heartily while standing in front of an industrial machine with complex machinery in the background, creating a sense of a bustling factory environment. His laughter is contagious, brightening the otherwise serious atmosphere of the workshop, where metal parts and tools are scattered across workbenches. The scene captures the engineer's genuine joy, offering a glimpse into the lighter moments within a typically intense work setting.]" }, { "from": "gpt", "value": "Normal. The engineer's face in the video is slightly blurry." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A Caspian tern soars gracefully through the air, its large size and robust build evident as it navigates the skies above a serene coastal landscape. With a striking orange-red bill that is thick and strong, this bird exemplifies power and agility. Its plumage is predominantly white, with a faint tinge of grey on the back, and its wings display black tips that contrast beautifully with the rest of its feathers. As it glides over the water, the tern's keen eyes search for prey, showcasing its hunting prowess against the backdrop of a calm sea and a clear blue sky.]" }, { "from": "gpt", "value": "Normal. The gull's wing proportions in the video are not well balanced." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A muscular man with a shaved head, wearing a black wetsuit, emerges from a pool on a space station, the water dripping from his body as he looks determined. The futuristic prison, with its high-tech security systems and alien guards, is visible in the background. The man, an inmate, uses his skills as a competitive swimmer to escape through the station's aquatic ventilation system, propelling himself through water-filled tunnels with powerful strokes. He reaches a critical junction, where he must hold his breath and navigate a complex series of underwater passages to avoid detection by the guards. Finally, he breaks the surface in an outdoor area, gasping for air and celebrating his hard-won freedom under the stars.]" }, { "from": "gpt", "value": "Normal. The person's arm is twisted, the hand is twisted, and the face is distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A police officer, wearing a full uniform complete with a badge, gun, and radio, is seen in a modern kitchen with gray walls, white cabinets, and stainless steel appliances. He is intently loading dishes into a dishwasher, his focus and precision reflecting his attention to detail, typically associated with his law enforcement duties. The kitchen counters are clean and clutter-free, except for a few dishware items awaiting their turn in the dishwasher. The officer's demeanor suggests a moment of domestic routine amidst his usual high-stakes professional life.]" }, { "from": "gpt", "value": "Bad. The character's face and hands are distorted in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist wearing a blue denim jacket and black cap, holding a paintbrush in his right hand, walks alongside a large elephant with intricate, colorful patterns painted on its body. Both the artist and the elephant stroll through a lush, green landscape filled with tall trees and vibrant vegetation. The elephant carries a canvas on its back, and the artist occasionally glances at it as they walk together in harmony. They are surrounded by the natural beauty of the environment, which complements the vivid artwork on the elephant's skin.]" }, { "from": "gpt", "value": "Bad. The elephant's walking motion in the video is uncoordinated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A visionary director, wearing a brown jacket, sits on a chair, intently operating a camera, capturing footage of a rhinoceros in a barren desert landscape, with a massive white hole explosion occurring in the distance. The rhinoceros, filmed in extreme close-up, slowly crawls across the sandy terrain, its textured skin and concentrated gaze highlighting its determination. The director, now seen from behind, continues to film the rhinoceros as it traverses the challenging environment, with the white hole explosion's brilliant light illuminating the scene, creating a surreal and captivating contrast between the earthbound creature and the cosmic event.]" }, { "from": "gpt", "value": "Bad. The person in the video is not shown." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged doctor, wearing a gray sweater over a white shirt, dark jeans, and black shoes, stands in his well-lit garage, surrounded by various tools and equipment. He is organizing his belongings, rearranging items on metal shelves, and sweeping the floor. The garage walls are painted a light color, and the space is cluttered yet functional. As he works, the doctor occasionally pauses to examine objects more closely, demonstrating care and attention to detail in his task of tidying up the garage environment.]" }, { "from": "gpt", "value": "Good. Basically correct." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A diligent male worker, clad in a green jacket, dark pants, and brown gloves, meticulously rakes a large pile of autumn leaves in the foreground, creating a tidy heap. Behind him, a vast expanse of fallen leaves covers the grassy area, extending into the distance. The scene is set in a spacious backyard surrounded by a wooden fence, with a row of mature trees framing the horizon. As the worker continues to rake, the pile of leaves grows larger, highlighting the laborious task at hand. Throughout the process, the worker maintains a focused demeanor, his movements methodical and efficient, embodying the essence of hard work and determination.]" }, { "from": "gpt", "value": "Bad. The character's hand is deformed, the face is distorted, and the rake in the hand is twisted and deformed." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A sleek, golden cheetah crawls under a low-hanging, silver photon sail, its eyes fixated ahead, muscles rippling beneath its spotted coat, set against a backdrop of a futuristic, metallic landscape. The sail, towering above, reflects beams of sunlight, creating an ethereal glow around the big cat. The cheetah, with its distinctive black tear markings and piercing green eyes, moves gracefully, its body positioned low to the ground, as if stalking prey, while the sail's advanced technology hints at a world beyond nature.]" }, { "from": "gpt", "value": "Good. Basically correct." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An engineer, clad in a grey hoodie, beige khakis, and brown boots, runs across a grassy field under a clear blue sky, holding a remote control. Nearby, a squirrel scampers along the ground, its bushy tail bouncing. In the distance, an AI-piloted drone hovers, capturing aerial footage. The engineer occasionally glances upwards, monitoring the drone's position as it flies, while the squirrel continues its journey among the trees. The scene is set in a suburban park, with houses visible in the background, highlighting the integration of technology and nature in everyday life.]" }, { "from": "gpt", "value": "Normal. The character's head is distorted, the arms are twisted, the legs are severely distorted, the squirrel's body is twisted, and the legs are twisted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A bearded male musician, wearing a brown hat and a blue shirt with suspenders, sits against a plain gray backdrop. He is knitting a long, cream-colored scarf with large needles, showcasing his dexterity. In another scene, he holds the growing scarf in his hands, his facial expression one of concentration. Next, he looks directly at the camera, his eyes conveying a sense of peace and contentment. Finally, he continues knitting, his fingers moving skillfully, set against the simple gray background that emphasizes his artistic endeavor.]" }, { "from": "gpt", "value": "Good. Basically no issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A visionary director, wearing a dark jacket and shirt, sits in a dimly lit room with neon accents, contemplating the concept of light-speed travel. He gently caresses a futuristic, spherical object, possibly a holographic model, as he ponders the intricacies of time and space. The camera captures his focused gaze and the intimate interaction with the sphere, symbolizing the connection between human creativity and the mysteries of the universe. The scene is set against a backdrop of high-tech, sci-fi inspired visuals, enhancing the theme of exploring the unknown.]" }, { "from": "gpt", "value": "Normal. The character in the video touches the spherical object with mostly normal movements, but there is some lack of coordination." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surprising scene set in a subtropical desert landscape, a police officer, clad in a tan uniform and wide-brimmed hat, is gently wiping the face of a curious capybara. The officer, showing unexpected tenderness, kneels beside the large rodent, which is sitting calmly on the sandy terrain, surrounded by sparse vegetation typical of arid environments. The interaction between the law enforcement figure and the capybara underlines an unusual yet touching moment in the vast, sun-baked wilderness.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A single kangaroo, identifiable by its reddish-brown fur, large muscular tail, and defined facial features, is seen jumping across a vast, arid landscape. The kangaroo bounds through the dry, sparsely vegetated terrain, its movements fluid and powerful. Its ears are alert, and its eyes are focused, conveying a sense of purpose and energy. The background is a vast, open space, punctuated by distant trees and a clear blue sky, emphasizing the kangaroo's natural habitat and the scale of its environment. The kangaroo's motion is captured in mid-air, showcasing the unique and iconic jumping ability that defines these marsupials.]" }, { "from": "gpt", "value": "Good. Basically no problem." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A large hippopotamus, identifiable by its gray skin and rotund body, is seen walking through a shallow river, creating ripples in the water with each step. Its small ears, round eyes, and wide mouth are clearly visible as it moves slowly through the riverbed. The scene is set against a backdrop of lush vegetation and a clear blue sky, with the water reflecting the greenery and the hippo's movements. The hippopotamus continues its leisurely walk, leaving a trail of disturbed water behind it, all captured in high definition to showcase the details of this magnificent creature in its natural habitat.]" }, { "from": "gpt", "value": "Normal. The sky was not generated as clear and blue." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vast, sandy desert under a cloudless blue sky, a lone brown camel slowly crawls across the dunes, its large, expressive eyes looking straight ahead. The animal's body is positioned low to the ground, its right front leg extended forward in a crawling motion, and its left hind leg bent beneath it. The texture of the camel's fur is detailed, showing the natural variations in color and the dryness of the skin. The sand dunes form dramatic curves and ridges, with soft shadows highlighting the contours created by the sun's light. In the distance, more dunes stretch to the horizon, emphasizing the camel's solitary journey through this expansive landscape.]" }, { "from": "gpt", "value": "Bad. The dromedary's body is twisted, its legs are twisted, and its face is distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A woman stands tall and poised, her blonde hair in a sleek bun, wearing a fitted black blazer over a white collared shirt, black trousers, and black heels. She is positioned in front of a large window with a city skyline visible in the background, her gaze directed towards the urban landscape. The scene is bathed in the warm glow of the setting sun, casting a golden hue over the woman and the cityscape, highlighting her silhouette and the texture of her attire. The contrast between the structured clothing and the fluidity of her stance conveys a sense of power and elegance.]" }, { "from": "gpt", "value": "Bad. The lady's head, hands, and feet in the video were generated abnormally." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [Two young children, a boy and a girl, are diligently cleaning a large fish tank positioned against a wall covered with a floral-patterned curtain. The boy, wearing a gray shirt and dark pants, uses a net to scoop debris from the water, while the girl, dressed in a yellow shirt and blue jeans, scrubs the glass with a small brush. Both children work attentively, surrounded by an array of aquarium supplies, including bags of gravel, fish food containers, and various cleaning tools. The tank is home to several colorful fish that swim tranquilly in the background, undisturbed by the children's careful efforts to maintain their environment.]" }, { "from": "gpt", "value": "Normal. The character's movements are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged man, donning a red plaid shirt, blue jeans, and a wide-brimmed straw hat, diligently sweeps his wooden porch with a long-handled broom, the early morning sun casting a warm glow over the rural scene. Beside him, a white dog rests lazily on the porch, enjoying the peaceful ambiance. The farmer pauses to wipe sweat from his brow, revealing a face lined with years of hard work and dedication to his land. He continues sweeping, the rhythmic sound of the broom against the wood providing a soothing soundtrack to the quiet countryside morning. Behind him, a lush green landscape stretches out, dotted with crops and farm buildings, a testament to his labor and connection to the earth.]" }, { "from": "gpt", "value": "Bad. The man's actions of sweeping the wooden porch with a long-handled broom are not continuous and smooth, there are abrupt changes; the man's action of stopping to wipe the sweat off his forehead is also not continuous and coordinated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A bearded man, presumably a writer, sits cross-legged on a lush, grassy surface, wearing a blue denim jacket and dark pants. He gently caresses a small, brown bear, who reciprocates with a loving nuzzle. The bear, adorned with a red bow tie, displays affection towards the man, who smiles warmly. Both are surrounded by a serene, verdant landscape, with the man's writing materials, including a notebook and pen, visible nearby. The scene conveys a tender moment between the writer and the bear, set against a peaceful outdoor backdrop.]" }, { "from": "gpt", "value": "Normal. No pen was generated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dimly lit room, a writer sits at a large wooden desk, adorned with a typewriter, a glass of whiskey, and a half-smoked cigar. Dressed in a brown leather jacket, black shirt, and jeans, he appears rugged and disheveled. Behind him, an open window reveals the gritty urban landscape of a city at night. Suddenly, the scene shifts to a close-up of a rhinoceros, its gray skin wrinkled and textured. The animal is being gently wiped down by a caregiver, who uses a cloth to clean its face, highlighting the creature's large size and docile nature. The juxtaposition of the writer and the rhinoceros creates a surreal and thought-provoking image.]" }, { "from": "gpt", "value": "Bad. The actions are not continuous, there are sudden changes." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a futuristic space station habitat, illuminated by soft, ambient lighting, a caring teacher, wearing a navy blue jumpsuit with silver accents, gently caresses the head of a young student seated at a desk. The student, dressed in a matching uniform, looks up with wide eyes, feeling comfort and support. Behind them, a large window reveals the stunning view of Earth, while various educational tools and floating objects fill the zero-gravity environment, creating an immersive learning atmosphere. The teacher's compassionate touch and the unique setting emphasize the nurturing bond between educator and pupil in this extraordinary space classroom.]" }, { "from": "gpt", "value": "Good. Basically no problem." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A woman wearing a yellow shirt and denim jeans is organizing a cluttered garage, illuminated by overhead lights. She sorts through items on a wooden table, her hands moving purposefully. Nearby, a brown dog sits attentively, watching her actions. Behind her, shelves filled with various objects and tools line the walls, indicating a space dedicated to storage and projects. The atmosphere is one of industriousness, as she focuses on tidying the space, creating order from chaos.]" }, { "from": "gpt", "value": "Bad. The light position in the video is incorrect, and the hands of the person and the face of the dog in the video are generated with abnormal distortions." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A rugged firefighter, clad in full gear and a yellow fire-resistant jacket, gently caresses a small meerkat standing on its hind legs. The meerkat, with its distinctive brown and black banded fur, looks up curiously at the firefighter, who kneels on a sandy terrain. Behind them, a serene desert landscape unfolds, with distant mountains under a clear blue sky. The firefighter's protective gear and the meerkat's natural habitat form an unlikely yet touching tableau of connection between a human and wildlife.]" }, { "from": "gpt", "value": "Bad. The meerkat's head and paws, as well as the human hand, were generated incorrectly in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical scene, a diverse group of characters gathers in a lush meadow under a clear blue sky. A male musician, dressed in a casual striped shirt, plays a ukulele, seated on a small wooden bench. Nearby, an animated gorilla, also wearing a striped shirt, energetically jumps into the air with excitement. The vibrant green grass and colorful flowers surrounding them add to the playful atmosphere, creating a harmonious blend of music, nature, and unexpected companionship.]" }, { "from": "gpt", "value": "Bad. The character's hands, face, and legs are distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A passionate female teacher, clad in a gray blazer, white blouse, and dark pants, walks purposefully across a bustling university campus, animatedly discussing the recent discovery of a habitable exoplanet. She gestures enthusiastically, engaging with her students who walk alongside her, their faces alight with curiosity and excitement. The backdrop shifts subtly from the interior of a modern lecture hall, filled with attentive students, to the outdoor campus, where the teacher continues her captivating lesson under the open sky, emphasizing the significance of the exoplanet find. Throughout the scene, the teacher's energy and dedication to teaching are evident, inspiring her students to explore the wonders of the universe.]" }, { "from": "gpt", "value": "Bad. The character's face in the video is slightly distorted, and the hands are twisted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical scene, a doctor, wearing a white lab coat and blue gloves, is seen wiping the nose of a friendly coyote who is sitting calmly on a checkered medical table. The room is brightly lit, with medical supplies neatly arranged on shelves in the background. The doctor's compassionate expression and the coyote's trusting demeanor create a heartwarming moment, highlighting an unusual yet tender interaction between human and wild animal in a clinical setting.]" }, { "from": "gpt", "value": "Bad. The character's face is distorted, the hands are twisted, the arms are deformed, and the coyote's head is contorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A single pigeon, predominantly gray with distinctive black markings, perches on a stone statue in the middle of a bustling city square. It has a full, rounded body, and its wings are slightly spread, revealing a flash of green and purple iridescence near the base. Its head, adorned with a small red beak, turns to face the camera, displaying its alertness and curiosity. The bird's legs, pink and sturdy, grip the statue firmly, while its red eyes, surrounded by yellow orbital rings, watch the surroundings keenly. The pigeon is set against a backdrop of urban life, with pedestrians, buildings, and traffic providing a sense of scale and context.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An artist dons a spacesuit and helmet, adorned with vibrant colors and patterns, resembling a solar sail, as they crawl across a barren lunar landscape, backlit by the Earth's glow. The suit, with its reflective, sail-like design, catches the sunlight, creating mesmerizing reflections and shadows. The artist moves slowly, emphasizing the suit's unique texture and the dramatic contrast between the suit and the moon's surface. This artistic portrayal on the moon's surface, with the distant Earth providing a serene backdrop, highlights the intersection of art and space exploration.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a unique fusion of scenes, a waiter in a crisp white shirt and black trousers serves a golden, perfectly roasted chicken on a rustic wooden table set amidst the lush greenery of a temperate rainforest. Moments later, the waiter, now donning a vibrant, traditional costume, gracefully dances among the towering trees, his movements harmonious with the serene environment. The chicken, now artfully presented on a platter, sits center stage as the waiter performs a captivating dance, his attire reflecting the rich cultural heritage associated with temperate rainforests, all under the canopy of majestic, moss-covered trees.]" }, { "from": "gpt", "value": "Bad. The person in the video has a twisted body, distorted face, twisted arms, twisted hands, and twisted legs." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a futuristic space station, a young female cashier, clad in a green jumpsuit with a name tag, works diligently at a market specializing in hydroponic produce. She crawls under the counter, surrounded by vibrant, healthy greens growing vertically against the back wall, illuminated by LED grow lights. The space agriculture setting is evident from the lush plants and the advanced technology used for cultivation. The cashier's actions are focused and purposeful, highlighting the unique challenges and innovations of farming in space, all within the confines of a bustling space station market.]" }, { "from": "gpt", "value": "Bad. The woman's arm is missing in the video, and her face and body are distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surreal, sci-fi setting, a large moose, adorned with a futuristic, glowing harness, stands calmly beside a colossal space elevator. The moose is being gently caressed by a figure dressed in a spacesuit, creating an unexpected yet tender moment between species. The vastness of space looms overhead, with stars and distant galaxies visible, contrasting with the intimate interaction on the ground. The space-suited individual continues to stroke the moose affectionately, while the majestic animal remains serene, its eyes reflecting curiosity and trust in this otherworldly encounter.]" }, { "from": "gpt", "value": "Normal. The character's movements in the video are basically normal, with slight disharmony." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A rugged man with a full beard, dressed in a green shirt and brown pants, is kneeling in a forest clearing. He gently wipes a small, injured porcupine with a white cloth, his face expressing deep concern and care. The porcupine, quills visible, lies quietly on a bed of moss and leaves, seemingly aware of the man's protective presence. The forest surroundings, with tall trees and dappled sunlight, provide a natural and peaceful setting for this tender interaction between man and wildlife.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a futuristic space setting, a renowned actor portrays a courageous character who masterminds a daring bank robbery on a distant planet. Dressed in a high-tech suit, the actor stealthily maneuvers through the alien bank, filled with luminescent crystals and advanced security systems. In a surprising twist, the actor gently caresses a small, friendly porcupine-like creature, which assists in the heist by using its quills to disable the security mechanisms. The scene combines elements of action, suspense, and unexpected tenderness in the vastness of outer space.]" }, { "from": "gpt", "value": "Bad. The actor's hands are distorted, the face is distorted, and the animal's head is distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A casually dressed woman, wearing a beige jacket and dark pants, walks her dog on a lead in a suburban neighborhood. She strolls along the pavement, passing by a row of neatly trimmed hedges and uniform houses with red roofs, creating a sense of domestic tranquility. The dog, a medium-sized brown and white spaniel, trots obediently beside her, its leash held gently in the woman's hand. As they walk, the afternoon sun casts long shadows, highlighting the peacefulness of the scene.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A distinguished film director, wearing a brown jacket, attentively wipes a gerbil, held gently in his hands, using a small towel, under the warm glow of studio lights. The gerbil, appearing relaxed and well-cared for, is nestled in the director's hands, which are meticulously cleaning it. The director's demeanor suggests a careful and nurturing approach to handling the small animal, creating a contrast between his authoritative role on set and this tender moment of care.]" }, { "from": "gpt", "value": "Normal. The director's hand, the hamster's head, and paws were generated abnormally in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A charismatic individual named Jay, donning a stylish gray jacket, a black turtleneck, and dark jeans, stands against a backdrop of city lights and a bridge, exuding confidence. He is captured in various poses, first looking directly at the viewer with a serious expression, then showing a side profile with a contemplative look. Jay then smiles warmly, showcasing his friendly demeanor, before striking a cool pose with his hands in his pockets, gazing into the distance. Throughout, the urban landscape behind him complements his modern style and enigmatic persona.]" }, { "from": "gpt", "value": "Good. Basically no problem." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A cheerful female teacher, wearing a green shirt and black pants, is seen in a classroom with a high ceiling, assembling a wooden table. The room is filled with natural light from large windows, illuminating the space. She is kneeling on the floor, carefully screwing in a leg of the table with a screwdriver, demonstrating a practical skill to her attentive students, who watch from a distance. The teacher's patience and focus are evident as she meticulously puts together the piece of furniture, providing a valuable learning experience in a bright and welcoming environment.]" }, { "from": "gpt", "value": "Normal. The teacher in the video assembled the wooden table basically normally, but there were slight movements that were not quite coordinated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vivid, close-up shot, a sleek, black king cobra emerges from the underbrush, its distinctive hood fully spread as it propels itself forward with surprising speed. Its body, covered in smooth scales, undulates gracefully over the forest floor, showcasing the powerful muscles beneath. The cobra's intense, amber eyes remain fixed ahead, conveying a sense of purpose and determination as it navigates the complex terrain with agility and precision. The lush greenery and dappled sunlight create a dynamic contrast against the snake's dark form, highlighting the primal energy of this extraordinary creature in motion.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A janitor, clad in a yellow safety vest and dark pants, sprints with urgency across a wooden bridge overlooking a geyser's steamy plume in the background. The scene shifts to show the janitor running from right to left, with the geyser's eruption seen erupting behind him, emphasizing his hurried pace. Next, he is depicted dashing down a stone pathway, with the geyser's intermittent eruptions providing a dramatic backdrop. Finally, the janitor is shown sprinting on a dirt path, with the geyser's majestic plumes in the distance, highlighting his perseverance against the geothermal landscape.]" }, { "from": "gpt", "value": "Bad. No wooden bridge was generated, and the geyser's water jet was inconsistent." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A stunning common pheasant, characterized by its vibrant plumage, is captured in a natural habitat, surrounded by lush green foliage. The male bird, with its distinctive red face, prominent white neck collar, and golden crest, struts confidently on the forest floor. Its iridescent green head and long, brownish-green tail feathers are highlighted as it moves through the underbrush. The setting showcases the pheasant's environment, featuring tall grasses and leafy plants, providing a rich, textured backdrop for this captivating display of avian beauty.]" }, { "from": "gpt", "value": "Good. Basically correct." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A lone coyote, with a gray and cream coat, runs across a sunlit prairie, its ears perked and tail held low. It bounds through the tall grass, leaving a slight dust cloud in its wake, indicating speed and purpose. The coyote's focused gaze and agile movements showcase its hunting prowess, as it navigates the open landscape under a clear blue sky. Moments later, it leaps over a small obstacle, demonstrating its agility and grace, all while maintaining its steady pace across the vast, golden plains.]" }, { "from": "gpt", "value": "Bad. The limbs of the coyote in the video have generation issues, with limbs suddenly disappearing or appearing in intermediate frames." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A wildlife park worker, clad in khaki shorts, a green shirt, and a wide-brimmed hat, interacts closely with a hippopotamus in a large, naturalistic enclosure filled with clear, blue water. The worker confidently swims alongside the hippo, which floats gracefully on its back, showcasing the unique bond between caretaker and animal in a peaceful, sunlit setting. Moments later, the worker is seen playfully engaging with the hippopotamus, which swims submerged in the water, its eyes closed in apparent bliss, highlighting the trust and connection they share in their aquatic environment.]" }, { "from": "gpt", "value": "Bad. The hippo's head and the character's arm are distorted in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A compassionate doctor, wearing a white lab coat, blue shirt, and khaki pants, gently caresses a small, brown squirrel nestled in her hands. The scene is set in a cozy, well-lit room with wooden floors, where the doctor, sporting a ponytail, glasses, and a warm smile, tenderly holds the squirrel. The focus is on the interaction between the caring medical professional and the tiny creature, emphasizing the nurturing bond formed in this peaceful environment.]" }, { "from": "gpt", "value": "Normal. The doctor's actions of petting the squirrel in the video are basically normal, with slight awkwardness." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a surreal, Mars-like landscape painted in shades of orange and red, a woman wearing a futuristic silver suit gently caresses a majestic bison, while a second bison stands nearby. Behind them, a man clad in a matching silver suit rides a robust, armored horse, creating a striking contrast against the barren, rocky terrain and the distant, hazy horizon. The woman's tender touch on the bison's fur and the man's confident posture on the horse convey a sense of harmony and adventure in this otherworldly setting.]" }, { "from": "gpt", "value": "Normal. The character's movements are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A compassionate, middle-aged female nurse sits on a wooden chair in a cozy, softly lit room, wearing a light blue scrubs top and black pants. She crochets a vibrant, multicolored blanket with yellow, orange, and red hues, her hands moving skillfully as she focuses on her craft. The room is filled with the warm glow of a standing lamp, casting a gentle light on the scene. Behind her, a large window reveals a lush, green outdoor landscape, providing a serene backdrop for her meditative activity.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A distinguished film director, donning a beige jacket, black t-shirt, and khaki pants, stands thoughtfully at a hot springs site, contemplating a unique shot. Suddenly, a rhinoceros emerges from the mist, leaping gracefully over a small pond within the springs. The director watches intently, camera in hand, ready to capture the extraordinary moment as the sun sets behind, casting a golden hue over the scene. The rhinoceros, now the central figure, jumps with surprising agility amidst the steamy waters, creating an unforgettable juxtaposition of nature and wildlife against the backdrop of the director's vision.]" }, { "from": "gpt", "value": "Bad. The video did not generate a small pond." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A large, formidable goose with striking orange beak and feet stands majestically on a grassy bank, its gray and white feathers ruffled by a gentle breeze. It stretches its neck upwards, revealing its full profile against the backdrop of a serene lake, dotted with lily pads and framed by lush greenery. The bird then lowers its head to preen its feathers, maintaining its composure as it interacts with its tranquil environment. Finally, it takes a step forward, ready to wade into the water, its bold presence commanding attention in this peaceful setting.]" }, { "from": "gpt", "value": "Normal. When the duck lowers its head, the head becomes deformed." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A majestic red deer leaps over a wooden fence in a lush, green field, its powerful legs and large frame displayed against a backdrop of trees and rolling hills. The deer, with an impressive set of antlers, is captured mid-jump, showing its agility and grace as it clears the barrier. The scene is set in a natural, countryside environment, with the deer's movement frozen in time, highlighting the beauty and strength of this wild animal.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A woman dressed in a panther costume, complete with a black leotard, detailed spots, and a lifelike mask, crawls on all fours across a dimly lit stage. The setting is enhanced by a spotlight illuminating her figure, emphasizing the fluidity and grace of her movements, akin to a real panther. The background remains shadowy, drawing attention to her performance as she embodies the essence of the wild animal through her crawling motion.]" }, { "from": "gpt", "value": "Bad. The woman's limbs and fingers are missing in the video, and her face is slightly distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An engineer, dressed in a formal black suit and tie, stands solemnly at a unique space-themed memorial service. He gently caresses the sleek, metallic casket adorned with constellations, reflecting a deep sense of loss and respect. The service takes place in a futuristic, well-lit room with large windows showcasing the vastness of space, complete with stars and galaxies. The engineer's emotional connection to the deceased is evident in his closed eyes and bowed head, while mourners in the background also dressed in black, pay their respects in a setting that combines elements of space exploration and human emotion.]" }, { "from": "gpt", "value": "Bad. The engineer's face and hands in the video are generated inconsistently." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A doctor, wearing a white lab coat and safety goggles, carefully crawls across a rugged, volcanic landscape, examining a unique rock formation. Surrounded by steam vents and volcanic rocks, the doctor, complete with a backpack and protective gear, crawls through a narrow crevice, demonstrating determination and focus in the harsh environment. The scene highlights the doctor's adventurous spirit and scientific curiosity, as they navigate the challenging terrain, crawling under low-hanging rock formations, all set against the dramatic backdrop of an active volcano.]" }, { "from": "gpt", "value": "Bad. The arms, hands, and head of the person in the video are distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A young woman wearing a yellow top and blue jeans is attentively cleaning a large, rectangular fish tank. She is seen smiling while scrubbing the glass with a sponge, ensuring a clear view of the colorful fish and underwater plants inside. The tank is illuminated by overhead lights, highlighting the woman's focused efforts. Her movements are gentle yet thorough, maintaining the aquatic environment's health and beauty. Throughout the process, the fish swim calmly around her, undisturbed by her presence, as she works diligently to keep their home pristine.]" }, { "from": "gpt", "value": "Bad. Facial distortion, hand distortion." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A vast school of silver sardines elegantly maneuvers through the clear blue ocean, their synchronized movements creating mesmerizing patterns as they navigate around a large, rocky coral reef structure. Sunlight filters through the water, illuminating the fish and highlighting their metallic scales against the vibrant coral and lush underwater flora. The sardines' fluid dynamics and the contrast between their silvery bodies and the colorful marine environment create a stunning visual display in the depths of the sea.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical scene, a cheerful nurse, clad in a traditional white uniform and cap, finds herself in an expansive landscape of lush farmlands under a clear blue sky. She shares an unexpected moment of joy with a playful otter, who leaps from her hands, its body arcing beautifully against the vibrant green backdrop of rolling fields and distant hills. The nurse's expression reflects pure delight as she watches the otter jump, encapsulating a surreal yet heartwarming encounter between human and nature.]" }, { "from": "gpt", "value": "Normal. The otter's jumping motion in the video is not smooth, and its limbs change abruptly." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man, embodying the essence of carefree summer days, walks along a pristine beach with white sand, wearing a vibrant floral shirt, khaki shorts, and brown leather sandals. He gently caresses the smooth, sun-warmed surface of a driftwood log, as the crystal-clear turquoise waters of the sea softly lap at the shore behind him. The serene beach landscape, with lush greenery and palm trees framing the horizon, complements the peaceful atmosphere. The man's relaxed posture and the idyllic beach setting evoke a sense of tranquility and connection with nature.]" }, { "from": "gpt", "value": "Bad. The man's face and hands in the video have generation issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a vast, cosmic library filled with floating bookshelves and galaxies in the background, a majestic lion, adorned with a golden collar and a serious expression, is seen wiping the dust off ancient books with a delicate paw. The shelves are stocked with books of all sizes, and the lion moves between them, engaged in the meticulous task of cleaning. Surrounded by the grandeur of space and knowledge, the lion continues its duty, highlighting the serene yet awe-inspiring setting.]" }, { "from": "gpt", "value": "Normal. The lion's movements are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A woman, clad in a simple black one-piece swimsuit, swims calmly in a vast, prehistoric lake, with a woolly mammoth standing watchfully nearby on the shore, its massive frame and shaggy coat a stark contrast against the serene water. The setting sun casts a warm glow over the scene, highlighting the woman's strokes and the mammoth's still form, creating a captivating juxtaposition of ancient and modern elements in a natural landscape.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An engineer, clad in a blue shirt and khaki pants, leaps into the air with arms outstretched against the vast desert landscape, showcasing his adventurous spirit. Moments later, he is captured in mid-jump, his silhouette framed by the golden hues of the dunes and the clear blue sky. Then, he lands on the sandy terrain, his posture reflecting a sense of accomplishment, with the desert's serene beauty serving as a dramatic backdrop. Finally, he jumps again, this time with a wide smile, embodying the joy of exploring the wilderness, all set against the stunning desert scenery.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A sparrowhawk soars majestically across a clear blue sky, its sharp eyes scanning the terrain below for any signs of movement. Its streamlined body, covered in a mottled brown and gray plumage, cuts through the air with effortless grace. The bird's hooked beak and powerful talons are on full display as it dives suddenly, plummeting towards the earth in pursuit of prey. In a burst of speed, it snatches a small rodent from the tall grass, its sharp claws closing around the struggling creature. The sparrowhawk then rises again into the heavens, carrying its prize back to its nest, a picture of strength and agility against the vastness of the sky.]" }, { "from": "gpt", "value": "Bad. The goshawk's movements in the video are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical setting, a well-dressed lawyer, wearing a dark suit, finds himself in a courtroom that resembles a jungle environment. Seated at a desk adorned with law books and vines, he attentively wipes a chimpanzee's face, which is sitting calmly beside him. The chimp, looking somewhat bewildered, is wearing a miniature suit and tie, adding to the surreal atmosphere of the scene. The lawyer's serious demeanor contrasts with the absurdity of the situation, creating a unique visual narrative.]" }, { "from": "gpt", "value": "Bad. The character's face is distorted, the hands are distorted, and the chimpanzee's face is distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A teenage boy wearing a red jacket and khaki pants is sprinting across a grassy field, his gaze fixed ahead with determination. As he runs, a small rabbit enters the frame from the bottom right corner, bounding alongside him. The student acknowledges the rabbit with a brief glance, but both continue their run, the student maintaining his focus and the rabbit keeping pace, against the backdrop of a clear blue sky and distant green hills.]" }, { "from": "gpt", "value": "Bad. The number of rabbits generated in the video is incorrect." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A distinguished film director, clad in a brown jacket, blue jeans, and a green shirt, takes a break from his usual endeavors to tackle an autumn chore. He rakes leaves in a residential backyard, surrounded by lush greenery and tall trees. The scene opens with him standing confidently, holding a rake, and looking intently at the camera. Next, he begins raking piles of golden leaves, his focused expression and body language conveying his determination. The camera then captures him pausing to rest, leaning on the rake, and smiling contentedly at his progress. Finally, he returns to work, skillfully maneuvering the rake to gather the last of the fallen foliage, all while the warm, natural tones of autumn provide a picturesque backdrop.]" }, { "from": "gpt", "value": "Bad. The director leaning on a rake mentioned in the caption does not appear in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A majestic male lion, distinguished by his thick, dark mane, powerful build, and tawny coat, sprints across a vast, arid savanna. His piercing yellow eyes and focused expression convey determination. The lion's muscular legs propel him forward in a burst of speed, stirring up clouds of dust beneath his paws. The landscape is dotted with scattered acacia trees and distant mountains, providing a rich, natural setting for this awe-inspiring display of raw power and grace.]" }, { "from": "gpt", "value": "Normal. The lion's walking movements in the video are not smooth and coordinated." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A cheerful Asian cashier, wearing a red apron over a white collared shirt, black pants, and black shoes, is seen mid-air, jumping with both feet off the ground, her ponytail bouncing behind her. The camera captures her from various angles, showing her enthusiasm and energy as she performs this unexpected action within the confines of her workplace, a modern convenience store with neatly arranged shelves and bright lighting.]" }, { "from": "gpt", "value": "Normal. The woman's movements in the video are not smooth." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A construction worker, clad in a yellow safety vest, blue jeans, and brown work boots, takes a break from his laborious tasks to sketch on a piece of paper, seated on a wooden beam at a construction site bustling with activity. Amidst the backdrop of unfinished buildings and heavy machinery, he focuses intently on his art, revealing a moment of unexpected creativity in the midst of a gritty urban environment. The scene captures his dedication to both his profession and his passion, offering a glimpse into the dual nature of his daily life.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A small, plump quail with brown and black speckled feathers stands prominently in the foreground, nestled among lush green foliage that includes leaves, ferns, and tall grasses. It faces left, its head turned slightly toward the camera, revealing its distinctive facial features, including a small beak and alert eyes. The bird's textured feathers are intricately detailed, and its legs are visible as it perches on the forest floor, surrounded by the vibrant flora.]" }, { "from": "gpt", "value": "Normal. The quail's feet in the video are generated with deformities." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A chimpanzee, characterized by its dark fur and physical strength, leaps powerfully from a low branch in a lush, green forest. The chimp, wearing a radio collar for tracking purposes, is seen mid-jump, demonstrating its agility and vitality. Its movement is captured in high definition, highlighting the details of its fur and the determination in its eyes. The chimpanzee propels itself forward, leaving the branch behind, showcasing its natural habitat and behaviors in a captivating, close-range shot.]" }, { "from": "gpt", "value": "Bad. The chimpanzee's mouth and hand generation in the video have issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [An attractive male actor, reminiscent of Michael B. Jordan, sits at a desk in a modern office space. Dressed in a casual white shirt, sleeves rolled up, and stylish glasses perched on his nose, he focuses intently on typing on a sleek, black keyboard. The desk is clutter-free, save for a single, elegant desk lamp casting a warm glow over the scene. Behind him, a large window frames a view of a bustling city skyline, hinting at the high-stakes narrative unfolding as his fingers dance across the keys.]" }, { "from": "gpt", "value": "Good. No obvious issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A doctor, identifiable by her stethoscope and medical attire, navigates a lunar buggy across the moon's desolate surface, which is marked by craters and strewn with rocks. The buggy's wheels kick up fine lunar dust as it traverses the barren landscape. The doctor's face, visible through the helmet of her spacesuit, reflects determination and focus. In a close-up shot, a small, resilient insect-like creature crawls slowly across the moon's rocky terrain, its tiny legs moving deliberately against the stark, gray backdrop. The buggy continues its journey, passing near the crawling creature, emphasizing the contrasting scales of movement and life on the moon.]" }, { "from": "gpt", "value": "Bad. The doctor did not appear in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A young, smiling nurse with a red stethoscope around her neck is seated on a grassy area, holding a small hedgehog in her hands, which looks curious and alert. Behind her, a vibrant, colorful depiction of a quantum leap phenomenon is visible, symbolized by swirling, dynamic patterns of light and color. In the background, a serene swimming pool with clear blue water and lush greenery can be seen, adding a peaceful contrast to the energetic visuals of the quantum leap. The nurse's attire includes a white uniform and name badge, while the hedgehog is depicted in natural detail, its spines clearly visible.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A musician, wearing a black jacket and blue jeans, gently caresses the head of a bison, which is lying down on a grassy field under a cloudy sky. The musician's left hand rests on the bison's snout, while the animal's eye is closed in apparent comfort. Both figures are set against a backdrop of green fields and distant mountains, with the bison's body partially obscured by the tall grass. The musician's right hand holds a harmonica, which hangs loosely by its strap, adding a musical element to the scene of connection between human and animal.]" }, { "from": "gpt", "value": "Bad. The person's hand is distorted in the video, and the buffalo's face is also distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [Amid the lush greenery of a tropical rainforest, twin toddlers, a boy and a girl, crawl on a mossy rock near a cascading waterfall, their clothing soaked from the water. The boy wears a red shirt and gray shorts, while the girl is dressed in a pink top and yellow shorts, both sporting wide smiles as they explore their watery surroundings. Their crawling movements contrast with the powerful rush of the waterfall, creating a dynamic scene of innocence and nature's grandeur. The camera captures this intimate moment, focusing on the children's expressions of wonder and joy as they interact with the natural world around them.]" }, { "from": "gpt", "value": "Bad. The faces of the boy and girl in the video are slightly blurred, the fingers are poorly generated, and the boy's body is distorted." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a dimly lit room, a focused doctor, dressed in a white lab coat, leans over a table, examining a specimen. On the floor below, a tiny mouse scurries across the wooden surface, navigating the shadows beneath the furniture. The scene alternates between the doctor's intense study and the mouse's cautious movements, highlighting the stark contrast between the clinical environment and the unexpected presence of the small creature.]" }, { "from": "gpt", "value": "Good. No obvious problems." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged doctor, wearing a white lab coat over a blue button-down shirt, sits in a modern office filled with medical equipment. He is holding a pen and papers, looking concerned as he coughs into his right hand. The scene is set indoors, with the doctor seated at a desk cluttered with medical files and supplies. Soft, ambient lighting highlights the professional yet somber atmosphere. The doctor's cough is noticeable, drawing attention to his discomfort amidst the sterile environment.]" }, { "from": "gpt", "value": "Normal. The doctor mentioned in the caption holding a pen and documents does not appear in the video." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man, clad in a brown jacket and khaki pants, stands in a lush, green forest clearing, gently wiping mud off the antlers of a majestic deer using a soft cloth. The deer, with large, inquisitive eyes, tolerates the man's ministrations, standing still as the man carefully cleans each tine of the impressive antlers. Sunlight filters through the canopy above, casting dappled shadows on the scene below, highlighting the tender interaction between the man and the wild creature.]" }, { "from": "gpt", "value": "Bad. The hand of the character in the video is generated incorrectly." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a cozy, rustic coffee shop filled with warm, golden light and the aroma of freshly brewed coffee, a young female cashier, wearing a green apron over a striped shirt, knits a vibrant red scarf during a lull in business. She sits behind the counter, surrounded by pastries, a cash register, and an espresso machine. Her focused expression and nimble fingers working the knitting needles capture the essence of making something beautiful by hand in a quaint, welcoming environment.]" }, { "from": "gpt", "value": "Normal. The character's actions of knitting a scarf in the video are basically normal, with only slight awkwardness." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man with short brown hair, neatly trimmed beard, and glasses stands in a modern office setting, wearing a dark gray blazer, white shirt, and jeans. He points his right index finger upward with a serious yet excited expression, indicating an idea or making a significant point. Behind him, a large window reveals a cityscape, adding depth and context to the scene. The office is sparsely decorated, featuring a simple desk and chair, which complements the minimalist aesthetic of the space.]" }, { "from": "gpt", "value": "Normal. The fingers of the hand appear deformed." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the motion issues in this video, focusing on the continuity and smoothness of actions, as well as their coherence and adherence to the laws of physics. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [In a whimsical, cartoon-like setting, a mongoose, dressed as a nurse in a white uniform, cap, and apron, is seen wiping the face of a young boy with a small towel. The boy, wearing a striped shirt, sits on a chair, looking up with wide eyes, as the animal nurse meticulously cleans him. The scene takes place in a room with white walls, and a painting of a ship hangs in the background, adding a touch of adventure to the otherwise ordinary setting.]" }, { "from": "gpt", "value": "Bad. The actions are not continuous, there are sudden changes." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A man with short brown hair, neatly trimmed beard, and glasses stands in a modern office setting, wearing a dark gray blazer, white shirt, and jeans. He points his right index finger upward with a serious yet excited expression, indicating an idea or making a significant point. Behind him, a large window reveals a cityscape, adding depth and context to the scene. The office is sparsely decorated, featuring a simple desk and chair, which complements the minimalist aesthetic of the space.]" }, { "from": "gpt", "value": "Normal. Fingers are deformed. The face is also deformed." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the semantic consistency issues in this video, focusing on the alignment between the text and the visual content. Specifically, point out any discrepancies regarding the subject (e.g., person or animal), quantity, color, scene description, style, and other relevant aspects. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A middle-aged male janitor, dressed in a gray sweater over a white collared shirt and khaki pants, pauses from his cleaning duties to take deep breaths, standing in a brightly lit, modern office hallway. He holds a mop in his right hand and rests his left hand on a mop bucket, while his eyes are closed, focusing on his breathing. The clean, minimalist surroundings emphasize the tranquility of the moment as he takes a moment to collect himself, his facial expression one of peaceful concentration.]" }, { "from": "gpt", "value": "Bad. The video does not show the man holding the mop with his right hand and resting his left hand on the mop bucket." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A fluffy, yellow chick with a small, red beak is seen crawling across a surface covered in hay, its tiny feet moving steadily as it explores its surroundings. The chick pauses momentarily, looking around with bright eyes, before continuing its crawl. The camera captures the chick from various angles, highlighting its downy feathers and the texture of the hay beneath it. Throughout the video, the chick remains the central focus, its curious movements and adorable appearance captivating the viewer's attention.]" }, { "from": "gpt", "value": "Normal. The chicken's claws in the video have issues." } ]

[ { "from": "human", "value": "<video>\nSuppose you are an expert in judging and evaluating the quality of AI-generated videos, please watch the frames of a given video and see the text prompt for generating the video.\nThen identify the fidelity issues in this video, assessing the realism of the content, including people, animals, and other objects. Highlight any problems such as missing limbs, deformed hands or faces, or other unrealistic elements. Provide a rating evaluation: Good, Normal, or Bad, along with a concise explanation supporting your assessment.\nYour task is provided as follows: Text Prompt: [A woman dressed in a flowing, red, full-length dress dances gracefully in the desert landscape at golden hour, with the warm, glowing sunlight illuminating her figure. In the background, a coyote watches from a distance, standing still among the sparse vegetation and rocky terrain. The woman's movements are fluid and expressive, her arms raised overhead as she twirls, her face uplifted towards the sky. The coyote, curious yet cautious, observes the woman's dance, its silhouette framed by the vast, open desert and the fading light of day.]" }, { "from": "gpt", "value": "Bad. The woman's face and limbs are missing in the video, as well as the wolf's face and limbs." } ]