From Cells to Sentences: An End-to-End Framework for Table Understanding

Deepak Vijaykeerthy, Arvind Agarwal
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:2458-2466, 2026.

Abstract

Real-world tables are messy: column headers are inconsistent, cells contain errors or missing values, and crucial information is scattered across multiple tables and documents. These issues cause even state-of-the-art language models to fail at seemingly simple questions. We present a robust framework for table understanding that explicitly handles these challenges through three coordinated mechanisms: structure-aware encoders that learn invariance to common corruptions, trainable slots that compress evidence to a fixed-size representation, and grounding modules that align each slot to supporting text passages. Unlike prior work that treats tables as flat text or relies on clean schemas, our approach maintains strong performance even under schema corruption and structural perturbations. Across eight benchmarks spanning question answering, fact verification, and text generation, we achieve the best performance among methods without external tools on five tasks and remain competitive with systems using much larger models or SQL executors. Under schema corruption and row/column permutations, our method degrades by less than 2 points, while baselines drop by up to 6-22 points, confirming that explicit denoising and grounding are essential for robust table understanding.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-vijaykeerthy26a, title = { From Cells to Sentences: An End-to-End Framework for Table Understanding }, author = {Vijaykeerthy, Deepak and Agarwal, Arvind}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {2458--2466}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/vijaykeerthy26a/vijaykeerthy26a.pdf}, url = {https://proceedings.mlr.press/v300/vijaykeerthy26a.html}, abstract = { Real-world tables are messy: column headers are inconsistent, cells contain errors or missing values, and crucial information is scattered across multiple tables and documents. These issues cause even state-of-the-art language models to fail at seemingly simple questions. We present a robust framework for table understanding that explicitly handles these challenges through three coordinated mechanisms: structure-aware encoders that learn invariance to common corruptions, trainable slots that compress evidence to a fixed-size representation, and grounding modules that align each slot to supporting text passages. Unlike prior work that treats tables as flat text or relies on clean schemas, our approach maintains strong performance even under schema corruption and structural perturbations. Across eight benchmarks spanning question answering, fact verification, and text generation, we achieve the best performance among methods without external tools on five tasks and remain competitive with systems using much larger models or SQL executors. Under schema corruption and row/column permutations, our method degrades by less than 2 points, while baselines drop by up to 6-22 points, confirming that explicit denoising and grounding are essential for robust table understanding. } }
Endnote
%0 Conference Paper %T From Cells to Sentences: An End-to-End Framework for Table Understanding %A Deepak Vijaykeerthy %A Arvind Agarwal %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-vijaykeerthy26a %I PMLR %P 2458--2466 %U https://proceedings.mlr.press/v300/vijaykeerthy26a.html %V 300 %X Real-world tables are messy: column headers are inconsistent, cells contain errors or missing values, and crucial information is scattered across multiple tables and documents. These issues cause even state-of-the-art language models to fail at seemingly simple questions. We present a robust framework for table understanding that explicitly handles these challenges through three coordinated mechanisms: structure-aware encoders that learn invariance to common corruptions, trainable slots that compress evidence to a fixed-size representation, and grounding modules that align each slot to supporting text passages. Unlike prior work that treats tables as flat text or relies on clean schemas, our approach maintains strong performance even under schema corruption and structural perturbations. Across eight benchmarks spanning question answering, fact verification, and text generation, we achieve the best performance among methods without external tools on five tasks and remain competitive with systems using much larger models or SQL executors. Under schema corruption and row/column permutations, our method degrades by less than 2 points, while baselines drop by up to 6-22 points, confirming that explicit denoising and grounding are essential for robust table understanding.
APA
Vijaykeerthy, D. & Agarwal, A.. (2026). From Cells to Sentences: An End-to-End Framework for Table Understanding . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:2458-2466 Available from https://proceedings.mlr.press/v300/vijaykeerthy26a.html.

Related Material