# VideoScore2: Think before You Score in Generative Video Evaluation

arXiv: https://arxiv.org/abs/2509.22799 · HTML: https://mlsanity.com/paper/2509.22799/

## Key facts

- Cited by 40 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 48 references that are in the map.
- First posted on arXiv 26 Sep 2025.

## Builds on (most cited first, 30 of 38)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning](https://mlsanity.com/paper/2307.04725/) (2023)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training](https://mlsanity.com/paper/2501.17161/) (2025)
- [Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators](https://mlsanity.com/paper/2303.13439/) (2023)
- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [LTX-Video: Realtime Video Latent Diffusion](https://mlsanity.com/paper/2501.00103/) (2024)
- [Latte: Latent Diffusion Transformer for Video Generation](https://mlsanity.com/paper/2401.03048/) (2024)
- [Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levels](https://mlsanity.com/paper/2312.17090/) (2023)
- [Video-R1: Reinforcing Video Reasoning in MLLMs](https://mlsanity.com/paper/2503.21776/) (2025)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives](https://mlsanity.com/paper/2211.04894/) (2022)
- [Open-Sora Plan: Open-Source Large Video Generation Model](https://mlsanity.com/paper/2412.00131/) (2024)
- [Improving Video Generation with Human Feedback](https://mlsanity.com/paper/2501.13918/) (2025)
- [Unified Reward Model for Multimodal Understanding and Generation](https://mlsanity.com/paper/2503.05236/) (2025)
- [VideoScore: Building Automatic Metrics to Simulate Fine-grained Human Feedback for Video Generation](https://mlsanity.com/paper/2406.15252/) (2024)
- [VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generation](https://mlsanity.com/paper/2503.06800/) (2025)
- [Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distribution](https://mlsanity.com/paper/2501.11561/) (2025)
- [Koala-36M: A Large-scale Video Dataset Improving Consistency between Fine-grained Conditions and Video Content](https://mlsanity.com/paper/2410.08260/) (2024)
- [VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models](https://mlsanity.com/paper/2403.06098/) (2024)
- [Q-Insight: Understanding Image Quality via Visual Reinforcement Learning](https://mlsanity.com/paper/2503.22679/) (2025)

## Built on by (most cited first, 30 of 40)

- [Speed by Simplicity: A Single-Stream Architecture for Fast Audio-Video Generative Foundation Model](https://mlsanity.com/paper/2603.21986/) (2026)
- [Video Generation Models Are Good Latent Reward Models](https://mlsanity.com/paper/2511.21541/) (2025)
- [Unified Personalized Reward Model for Vision Generation](https://mlsanity.com/paper/2602.02380/) (2026)
- [VQQA: An Agentic Approach for Video Evaluation and Quality Improvement](https://mlsanity.com/paper/2603.12310/) (2026)
- [PhyGround: Benchmarking Physical Reasoning in Generative World Models](https://mlsanity.com/paper/2605.10806/) (2026)
- [Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey](https://mlsanity.com/paper/2510.01925/) (2025)
- [Benchmarking Scientific Understanding and Reasoning for Video Generation using VideoScience-Bench](https://mlsanity.com/paper/2512.02942/) (2025)
- [VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction](https://mlsanity.com/paper/2602.13294/) (2026)
- [WorldReasonBench: Human-Aligned Stress Testing of Video Generators as Future World-State Predictors](https://mlsanity.com/paper/2605.10434/) (2026)
- [CineDance: Towards Next-Generation Multi-Shot Long-Form Cinematic Audio-Video Generation](https://mlsanity.com/paper/2606.09639/) (2026)
- [A Benchmark for Omni-Modal Reasoning in Long Videos](https://mlsanity.com/paper/2512.16978/) (2025)
- [SPIRAL: Self-Evolving Action-Conditioned Video Generation via Reflective Planning Agents](https://mlsanity.com/paper/2603.08403/) (2026)
- [RewardHarness: Self-Evolving Agentic Post-Training](https://mlsanity.com/paper/2605.08703/) (2026)
- [CaC: Advancing Video Reward Models via Hierarchical Spatiotemporal Concentrating](https://mlsanity.com/paper/2605.11723/) (2026)
- [SwitchCraft: Training-Free Multi-Event Video Generation with Attention Controls](https://mlsanity.com/paper/2602.23956/) (2026)
- [Inference-Time Scaling for Joint Audio-Video Generation](https://mlsanity.com/paper/2606.03183/) (2026)
- [Thinking with Frames: Generative Video Distortion Evaluation via Frame Reward Model](https://mlsanity.com/paper/2601.04033/) (2026)
- [MMPhysVideo: Physically Plausible Video Generation Through Joint RGB-Perception Modeling](https://mlsanity.com/paper/2604.02817/) (2026)
- [How Far Are Video Models from True Multimodal Reasoning?](https://mlsanity.com/paper/2604.19193/) (2026)
- [Think, then Score: Decoupled Reasoning and Scoring for Video Reward Modeling](https://mlsanity.com/paper/2605.05922/) (2026)
- [Sci-VBench: Evaluating Knowledge- and Reasoning-Intensive Video Generation in Science Domains](https://mlsanity.com/paper/2608.09873/) (2026)
- [FIRM-Video: Check Before You Score for Reliable Text-to-Video Reward Modeling](https://mlsanity.com/paper/2608.21839/) (2026)
- [Early Failure Detection and Intervention in Video Diffusion Models](https://mlsanity.com/paper/2603.14320/) (2026)
- [AnimationBench: Are Video Models Good at Character-Centric Animation?](https://mlsanity.com/paper/2604.15299/) (2026)
- [LoViF 2026 The First Challenge on Holistic Quality Assessment for 4D World Model (PhyScore)](https://mlsanity.com/paper/2605.05187/) (2026)
- [SpecLoR: Spectral Lookahead Rectification for Motion-Coherent Text-to-Video Generation](https://mlsanity.com/paper/2606.11969/) (2026)
- [RoboGaze: Evaluating Robot World Models via Structured Vision-Language Analysis](https://mlsanity.com/paper/2606.28385/) (2026)
- [Distilling Physical Priors into Streaming World Models](https://mlsanity.com/paper/2608.07981/) (2026)
- [RewardVerse: Rubric-Guided Policy Optimization for Video Reward Modeling](https://mlsanity.com/paper/2609.22947/) (2026)
- [Generative Action Tell-Tales: Assessing Human Motion in Synthesized Videos](https://mlsanity.com/paper/2512.01803/) (2025)

## Abstract (arXiv)

Recent advances in text-to-video generation have produced increasingly realistic and diverse content, yet evaluating such videos remains a fundamental challenge due to their multi-faceted nature encompassing visual quality, semantic alignment, and physical consistency. Existing evaluators and reward models are limited to single opaque scores, lack interpretability, or provide only coarse analysis, making them insufficient for capturing the comprehensive nature of video quality assessment. We present VideoScore2, a multi-dimensional, interpretable, and human-aligned framework that explicitly evaluates visual quality, text-to-video alignment, and physical/common-sense consistency while producing detailed chain-of-thought rationales. Our model is trained on a large-scale dataset VideoFeedback2 containing 27,168 human-annotated videos with both scores and reasoning traces across three dimensions, using a two-stage pipeline of supervised fine-tuning followed by reinforcement learning with Group Relative Policy Optimization (GRPO) to enhance analytical robustness. Extensive experiments demonstrate that VideoScore2 achieves superior performance with 44.35 (+5.94) accuracy on our in-domain benchmark VideoScore-Bench-v2 and 50.37 (+4.32) average performance across four out-of-domain benchmarks (VideoGenReward-Bench, VideoPhy2, etc), while providing interpretable assessments that bridge the gap between evaluation and controllable generation through effective reward modeling for Best-of-N sampling. Project Page: https://tiger-ai-lab.github.io/VideoScore2/