EUVIMEDEUVIMED entdecken
Crossref

Look as You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning

Shuochen Liu, Pengfei Luo, Chao Zhang, Yuhao Chen, Haotian Zhang, Qi Liu, Xin Kou, Tong Xu, Enhong Chen

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (38) · 32159-32167

DOI: 10.1609/aaai.v40i38.40488 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Shuochen Liu und 8 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aiming to identify precise evidence sources from visual documents, visual evidence attribution for visual document retrieval–augmented generation (VD-RAG) ensures reliable and verifiable predictions from vision-language models (VLMs) in multimodal question answering. Most existing methods adopt end-to-end training to facilitate intuitive answer verification. However, they lack fine-grained supervision and progressive traceability throughout the reasoning process. In this paper, we introduce the Chain-of-Evidence (CoE) paradigm for VD-RAG. CoE unifies Chain-of-Thought (CoT) reasoning and visual evidence attribution by grounding reference elements in reasoning steps to specific regions with bounding boxes and page indexes. To enable VLMs to generate such evidence-grounded reasoning, we propose Look As You Think (LAT), a reinforcement learning framework that trains models to produce verifiable reasoning paths with consistent attribution. During training, LAT evaluates the attribution consistency of each evidence region and provides rewards only when the CoE trajectory yields correct answers, encouraging process-level self-verification. Experiments on vanilla Qwen2.5-VL-7B-Instruct with Paper‑ and Wiki‑VISA benchmarks show that LAT consistently improves the vanilla model in both single- and multi-image settings, yielding average gains of 8.23% in soft exact match (EM) and 47.0% in IoU@0.5. Meanwhile, LAT not only outperforms the supervised fine-tuning baseline, which is trained to directly produce answers with attribution, but also exhibits stronger generalization across domains.

Zitieren

APA

Liu, S., Luo, P., Zhang, C., Chen, Y., Zhang, H., Liu, Q., Kou, X., Xu, T., & Chen, E. (2026). Look as You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(38), 32159-32167. https://doi.org/10.1609/aaai.v40i38.40488

Vancouver

Liu S, Luo P, Zhang C, Chen Y, Zhang H, Liu Q, et al. Look as You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(38):32159-32167. doi:10.1609/aaai.v40i38.40488

BibTeX
@article{Liu2026,
  author = {Liu, Shuochen and Luo, Pengfei and Zhang, Chao and Chen, Yuhao and Zhang, Haotian and Liu, Qi and Kou, Xin and Xu, Tong and Chen, Enhong},
  title = {Look as You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {38},
  pages = {32159--32167},
  doi = {10.1609/aaai.v40i38.40488}
}
Kostenlos starten →