PhyRubric: Rubric-Grounded Diagnostic Evaluation of Physical Commonsense in Text-to-Video Generation
cs.CV, cs.AI
Submitted: 2026-11-10
Updated: 2026-11-10
Code: https://github.com/chenyil6/phyRubric
Terminology
Sources
- Cosmos World Foundation Model Platform for Physical AI
- HealthBench: Evaluating Large Language Models Towards Improved Human Health
- Qwen3-VL Technical Report
- MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark
- VideoCrafter1: Open Diffusion Models for High-Quality Video Generation
- From Holistic Evaluation to Structured Criteria: Rubrics Across the Evolving LLM Landscape
- TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation
- T2VPhysBench: A First-Principles Benchmark for Physical Consistency in Text-to-Video Generation
- LTX-Video: Realtime Video Latent Diffusion
- LTX-2: Efficient Joint Audio-Visual Foundation Model
- VideoScore2: Think before You Score in Generative Video Evaluation
- LoRA: Low-Rank Adaptation of Large Language Models
- How Far is Video Generation from World Model: A Physical Law Perspective
- PhyGround: Benchmarking Physical Reasoning in Generative World Models
- Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model
- Evaluating and Predicting Distorted Human Body Parts for Generated Images
- Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generation
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- From Sora What We Can See: A Survey of Text-to-Video Generation
- Towards Accurate Generative Models of Video: A New Metric & Challenges
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs