用正确答案反推最优删表路径，提升表格问答准确率
Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search
- 用真实查询路径监督并行搜索删表方案
- 平均准确率达73.5%，领先基线3.2个百分点
- 适合需要精准表格推理的场景
表格问答（TableQA）受益于表格剪枝技术，该技术通过剔除冗余单元格提取紧凑子表，以简化下游推理。然而，现有方法通常依赖不可靠的批评信号进行顺序修订，常遗漏关键答案数据。为此，我们提出TabTrim框架，将表格剪枝从顺序修订转变为基于黄金剪枝轨迹的并行搜索。TabTrim利用黄金SQL查询执行过程中的中间子表生成黄金剪枝轨迹，并训练剪枝器与验证器使每一步剪枝结果对齐该轨迹。推理时，TabTrim并行探索多个候选剪枝路径，识别最优子表。大量实验表明，TabTrim在多种表格推理任务中达到领先性能：TabTrim-8B平均准确率达73.5%，超越最强基线3.2%，其中WikiTQ上达79.4%，TableBench上达61.2%。
原文摘要 · Abstract (English)
Table Question Answering (TableQA) benefits significantly from table pruning, which extracts compact sub-tables by eliminating redundant cells to streamline downstream reasoning. However, existing pruning methods typically rely on sequential revisions driven by unreliable critique signals, often failing to detect the loss of answer-critical data. To address this limitation, we propose TabTrim, a novel table pruning framework which transforms table pruning from sequential revisions to gold trajectory-supervised parallel search. TabTrim derives a gold pruning trajectory using the intermediate sub-tables in the execution process of gold SQL queries, and trains a pruner and a verifier to make the step-wise pruning result align with the gold pruning trajectory. During inference, TabTrim performs parallel search to explore multiple candidate pruning trajectories and identify the optimal sub-table. Extensive experiments demonstrate that TabTrim achieves state-of-the-art performance across diverse tabular reasoning tasks: TabTrim-8B reaches 73.5% average accuracy, outperforming the strongest baseline by 3.2%, including 79.4% on WikiTQ and 61.2% on TableBench.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。