FIRM-Video: Check Before You Score for Reliable Text-to-Video Reward Modeling

  • 类型:arxiv
  • 标识:2608.21839
  • 链接:https://arxiv.org/abs/2608.21839
  • 主分类:multimodal
  • 形态:method
  • 被引:1
  • 被引来源:Semantic Scholar
  • S2被引:1
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:FHM-Video, a unified checklist-driven data construction framework based on a check-before-score principle, is introduced, which achieves the best overall MAE on FIRM-Video-Bench while consistently delivering the highest VBench Total, Quality, and Semantic Scores in Best-of-8 sampling across three video generators.
  • OpenAlex ID:W7204168750
  • OpenAlex DOI:10.48550/arxiv.2608.21839
  • DOI:10.48550/arxiv.2608.21839
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://arxiv.org/pdf/2608.21839
  • OpenAlex更新:2026-09-01
  • 副分类:evaluation
  • 待LLM分类:否
  • 标题中文:FIRM-Video:先核查再评分,实现可靠的 text-to-video 奖励建模
  • TLDR中文:本文提出 FHM-Video,一种基于 check-before-score 原则的、由 checklist 驱动的统一数据构建框架,在 FIRM-Video-Bench 上取得最佳综合 MAE,并在三种视频生成器的 Best-of-8 采样中始终获得最高的 VBench Total、Quality 和 Semantic Score。
  • 来源文件:
  • /inbox/tom/_candidates/2026-08-27-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]