arXiv cs.AI· Dunhan Li, Leilei Gan, Lirong Gao, Wenyan Fan, Yongwei Wang, Sheng Guo, Fei Wu·· 4 小时前AI 评分29
EviDARL:以证据为中心的双智能体交替强化学习框架,提升多表问答推理
EviDARL: Evidence-Centric Dual-Agent Reasoning via Alternating Reinforcement Learning for Multi-Table Question Answering
AI 导读
EviDARL 提出以证据为中心的双智能体框架 EviDA,由证据构建智能体与推理回答智能体组成,前者在完整表格上生成并执行表格操作程序,产出问题专属证据与可验证中间轨迹。EviDARL 结合智能体细粒度奖励与参数高效交替强化学习,轮流优化单个智能体并固定另一个,实现证据构建与答案推理的双向适配。在 MMQA 和 TQA-Bench 上,该方法在受限模型与训练预算下显著优于开源模型及多表推理基线。
正文 · 原文
Abstract:Despite advances in LLM-based table reasoning, most studies focus on single-table settings. Multi-table QA requires multi-hop reasoning over complex table relationships and evidence dispersed across long contexts, often without verifiable intermediate traces. To address these challenges, we introduce EviDA, an evidence-centric dual-agent framework comprising an Evidence Construction Agent and a Reasoning and Answering Agent. The Evidence Construction Agent generates and executes table-manipulation programs over complete tables to produce question-specific evidence and verifiable intermediate traces, simplifying subsequent reasoning. The Reasoning and Answering Agent then performs multi-hop reasoning over this evidence to derive final answers. We further introduce EviDARL, which combines agent-specific fine-grained rewards with parameter-efficient alternating reinforcement learning to improve collaboration between the agents. By optimizing each agent in turn while holding the other fixed, EviDARL enables bidirectional adaptation between evidence construction and answer reasoning. Experiments on MMQA and TQA-Bench demonstrate that EviDARL substantially outperforms competitive open-model and multi-table reasoning baselines under constrained model and training budgets.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.22310 [cs.AI] |
| (or arXiv:2611.22310v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.22310 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Dunhan Li [view email]
[v1]
Tue, 17 Nov 2026 07:07:41 UTC (2,015 KB)
来源:arXiv cs.AI · arxiv.org