用可追溯的物理验证系统，自动识别视频生成中的真实物理错误。
VeriPhy: Agentic Physical Reasoning for World Model Evaluation and Refinement
- 先用文本规划生成物理规则和执行计划，再通过冻结专家模块逐帧验证。
- 在149个视频上检出228处物理错误，远超现有方法，且每条结论有完整证据链。
- 适合需要可审计、可反馈的视频生成质量评估场景，如安全关键应用。
生成视频的视觉流畅性不等于物理可靠性，单一质量评分无法指出具体违反的物理规律或出错时刻。我们提出VeriPhy，一个可审计的物理验证系统：仅用文本规划器将提示转化为带类型的物理义务和静态验证的执行计划，无需观察任何画面。执行时，观测仅触发已声明的冻结低层专家（如分割、追踪、计数、十一种类型物理测量、深度、OCR、音频事件检测）。每个动作返回携带溯源信息的证据记录，其内容若可用，则为类型化测量或显式标记的学习状态。类型化解析器与固定组合逻辑将有效记录映射为三值状态（支持、矛盾或未知），以可信赖的‘合理、不合理、中立’形式呈现，确保每条结论均可追溯至原始证据。评估基于1,500段人类标注错误记录的语料库，精确定位生成失败在提示、空间与时间上的位置。在包含304条错误记录的149段核心数据集上，VeriPhy成功识别228条，优于已有问题分解评估器的164条；虽召回率与直接提示单体模型（222条）接近，但区别在于每个决策保留证据记录与溯源，实现逐条可审计，并可作为批评意见反写回生成过程的接口。
原文摘要 · Abstract (English)
Visual fluency in generated video does not imply physical reliability, and a scalar quality score alone is incapable of indicating the obligation a clip violates or the moment it fails. We present VeriPhy, an auditable physical-verification system in which a text-only planner compiles the prompt into typed physical obligations and a statically validated execution plan before any frame is observed. During execution, observations gate and scope only declared calls to frozen low-level experts (e.g., segmentation and tracking, counting, eleven typed physical measurements over the resulting tracks, depth, OCR, and audio-event detection). Each action returns a provenance-carrying evidence record whose payload, when usable, is either a typed measurement or an explicitly tagged learned state. Typed resolvers and fixed composition map usable records to a three-valued state (supported, contradicted, or unknown, surfaced as plausible, implausible, or abstain) with full provenance, so that every verdict is traceable to the evidence that produced it. We anchor evaluation in a 1,500-clip corpus of human-annotated flaw records that localize real generation failures in prompt reference, space, and time. On a 149-clip core carrying 304 such records, VeriPhy accounts for 228, against 164 for a published question-decomposition evaluator given the same clips and the same claims. Recall alone does not separate it from prompting the same backbone monolithically, which reaches 222; what separates them is that each decision retains its evidence record and provenance, making the traces auditable one verdict at a time and usable as the interface through which a critic verdict could be written back into generation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。