SAGE: From Direct Answering to Evidence-Grounded Inference for Chinese Ancient Document Understanding

2026-08-24

AI Summary

为解决古文理解中LVLMs生成答案过于自信且依据不足的问题，提出SAGE框架，通过多代理协调进行基于证据的推理，实验证明其优于直接作答模型。

Abstract

Chinese ancient document understanding demands complex visual, linguistic, and historical reasoning. Current Large Vision-Language Models (LVLMs) typically rely on an opaque, single-pass generation paradigm, often producing overconfident and weakly grounded responses. To address this, we propose SAGE, an evidence-grounded multi-agent framework that reformulates Chinese ancient document understanding as evidence-grounded inference rather than direct answer generation. SAGE coordinates specialized agents for task-aware planning, tool-mediated evidence acquisition, claim-level verification, and bounded replanning under a constrained shared-state runtime. This design supports bounded evidence seeking, answer revision, and abstention when grounding is insufficient. Experiments on the AncientDoc benchmark show that SAGE consistently outperforms matched direct-answering baselines across three LVLM backbones. Remarkably, SAGE with Qwen3.5-9B surpasses much larger monolithic LVLMs on most evaluated metrics, highlighting the importance of structured, evidence-grounded inference beyond model scaling.

Problem

• Chinese ancient document understanding
• Large Vision-Language Models (LVLMs)
• evidence-grounded inference

Innovation

• evidence-grounded inference
• multi-agent framework
• Chinese ancient document understanding
• task-aware planning