@inproceedings{zhang-etal-2026-tore,
title = "{TOR}e: Rehearsal-Free Continual Learning for Table {QA} via Operation Recipe Reconstruction",
author = "Zhang, Jinyu and
Liu, Ruiheng and
Zhang, Yu",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1320/",
pages = "30069--30095",
ISBN = "979-8-89176-459-0",
abstract = "Continual learning for Table Question Answering (TableQA-CL) aims to train models on sequential tasks with limited supervision per task. Existing methods either rely on replaying historical data or assume ideal conditions like gold task boundaries, limiting real-world applicability. We propose TORe (Table Operation Recipe Reconstruction), a rehearsal-free framework that reconstructs historical knowledge through schema-aware reasoning patterns. TORe extracts abstract Operation Recipes from table reasoning processes, identifies fine-grained knowledge deviations between current and past tasks, and employs execution-verified LLM synthesis to generate pseudo-examples reflecting both current and historical reasoning patterns. Experiments on two TableQA-CL benchmarks and multiple backbone models show that TORe consistently improves average accuracy over strong rehearsal-free baselines, remains competitive with replay-based methods, and maintains robust performance under different task orders."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-tore">
<titleInfo>
<title>TORe: Rehearsal-Free Continual Learning for Table QA via Operation Recipe Reconstruction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jinyu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruiheng</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Continual learning for Table Question Answering (TableQA-CL) aims to train models on sequential tasks with limited supervision per task. Existing methods either rely on replaying historical data or assume ideal conditions like gold task boundaries, limiting real-world applicability. We propose TORe (Table Operation Recipe Reconstruction), a rehearsal-free framework that reconstructs historical knowledge through schema-aware reasoning patterns. TORe extracts abstract Operation Recipes from table reasoning processes, identifies fine-grained knowledge deviations between current and past tasks, and employs execution-verified LLM synthesis to generate pseudo-examples reflecting both current and historical reasoning patterns. Experiments on two TableQA-CL benchmarks and multiple backbone models show that TORe consistently improves average accuracy over strong rehearsal-free baselines, remains competitive with replay-based methods, and maintains robust performance under different task orders.</abstract>
<identifier type="citekey">zhang-etal-2026-tore</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1320/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>30069</start>
<end>30095</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TORe: Rehearsal-Free Continual Learning for Table QA via Operation Recipe Reconstruction
%A Zhang, Jinyu
%A Liu, Ruiheng
%A Zhang, Yu
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zhang-etal-2026-tore
%X Continual learning for Table Question Answering (TableQA-CL) aims to train models on sequential tasks with limited supervision per task. Existing methods either rely on replaying historical data or assume ideal conditions like gold task boundaries, limiting real-world applicability. We propose TORe (Table Operation Recipe Reconstruction), a rehearsal-free framework that reconstructs historical knowledge through schema-aware reasoning patterns. TORe extracts abstract Operation Recipes from table reasoning processes, identifies fine-grained knowledge deviations between current and past tasks, and employs execution-verified LLM synthesis to generate pseudo-examples reflecting both current and historical reasoning patterns. Experiments on two TableQA-CL benchmarks and multiple backbone models show that TORe consistently improves average accuracy over strong rehearsal-free baselines, remains competitive with replay-based methods, and maintains robust performance under different task orders.
%U https://aclanthology.org/2026.emnlp-main.1320/
%P 30069-30095
Markdown (Informal)
[TORe: Rehearsal-Free Continual Learning for Table QA via Operation Recipe Reconstruction](https://aclanthology.org/2026.emnlp-main.1320/) (Zhang et al., EMNLP 2026)
ACL