Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Guo, Yu, Ye, Shenghao, Chen, Shuangwu, Wen, Zijian, Zhang, Tao, Bai, Qirui, Jin, Dong, Hou, Yunpeng, He, Huasen, Yang, Jian, Tan, Xiaobin
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916023260676096
author Guo, Yu
Ye, Shenghao
Chen, Shuangwu
Wen, Zijian
Zhang, Tao
Bai, Qirui
Jin, Dong
Hou, Yunpeng
He, Huasen
Yang, Jian
Tan, Xiaobin
author_facet Guo, Yu
Ye, Shenghao
Chen, Shuangwu
Wen, Zijian
Zhang, Tao
Bai, Qirui
Jin, Dong
Hou, Yunpeng
He, Huasen
Yang, Jian
Tan, Xiaobin
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2601_03851
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search
Guo, Yu
Ye, Shenghao
Chen, Shuangwu
Wen, Zijian
Zhang, Tao
Bai, Qirui
Jin, Dong
Hou, Yunpeng
He, Huasen
Yang, Jian
Tan, Xiaobin
Computation and Language
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.
title Rethinking Table Pruning in TableQA: From Sequential Revisions to Gold Trajectory-Supervised Parallel Search
topic Computation and Language
url https://arxiv.org/abs/2601.03851