DocVAL：用于基于文档的视觉问答的验证链式思维蒸馏
DocVAL: Validated Chain-of-Thought Distillation for Grounded Document VQA
AI总结：
提出DocVAL框架，通过验证链式思维蒸馏将大型教师模型的空间推理能力转移到紧凑学生模型，结合规则验证器和两阶段训练，在文档VQA任务上提升定位性能。
摘要：
文档视觉问答要求模型不仅正确回答问题，还要在复杂文档布局中精确定位答案。大型视觉语言模型（VLM）具有强大的空间定位能力，但其推理成本和延迟限制了实际部署。紧凑型VLM更高效，但在标准微调或蒸馏下常出现显著的定位退化。为解决这一问题，我们提出DocVAL，一种验证链式思维（CoT）蒸馏框架，将显式空间推理从大型教师模型转移到紧凑、可部署的学生VLM。DocVAL结合了（1）教师生成的空间CoT监督，（2）基于规则的双模式验证器，过滤低质量训练信号并提供细粒度像素级纠正反馈，以及（3）验证驱动的两阶段训练过程与迭代细化。文本检测仅作为训练时的监督和验证脚手架，使得最终学生模型在推理时作为纯VLM运行，无需OCR或检测。在多个文档理解基准上，DocVAL相比可比的紧凑VLM持续提升高达6-7个ANLS点。我们进一步引入平均精度（mAP）作为文档问答的定位指标，并在此新评估下报告了强大的空间定位性能。我们发布了95K验证器验证的CoT轨迹，并表明高质量、验证过的监督比扩展未过滤数据更有效，实现了高效且可信的文档定位。代码/数据：https://github.com/ahmad-shirazi/DocVAL
英文摘要：
Document visual question answering requires models not only to answer questions correctly, but also to precisely localize answers within complex document layouts. While large vision-language models (VLMs) achieve strong spatial grounding, their inference cost and latency limit real-world deployment. Compact VLMs are more efficient, but they often suffer substantial localization degradation under standard fine-tuning or distillation. To address this gap, we propose DocVAL, a validated chain-of-thought (CoT) distillation framework that transfers explicit spatial reasoning from large teacher models to compact, deployable student VLMs. DocVAL combines (1) teacher-generated spatial CoT supervision, (2) a rule-based dual-mode validator that filters low-quality training signals and provides fine-grained, pixel-level corrective feedback, and (3) a validation-driven two-stage training procedure with iterative refinement. Text detection is used only as training-time scaffolding for supervision and validation, enabling the final student to operate as a pure VLM without OCR or detection at inference. Across multiple document understanding benchmarks, DocVAL yields consistent improvements of up to 6-7 ANLS points over comparable compact VLMs. We further introduce mean Average Precision (mAP) as a localization metric for document question answering and report strong spatial grounding performance under this new evaluation. We release 95K validator-verified CoT traces and show that high-quality, validated supervision is more effective than scaling unfiltered data, enabling efficient and trustworthy document grounding. Code/Data: https://github.com/ahmad-shirazi/DocVAL
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
