Towards Better Understanding Table Instruction Tuning: Decoupling the Effects from Data versus Models

ACL ARR 2025 February Submission143 Authors

Abstract:

Recent advances in natural language processing have leveraged instruction tuning to enhance Large Language Models (LLMs) for table-related tasks.
However, previous works train different base models with different training data, lacking an apples-to-apples comparison across the result table LLMs.
To address this, we fine-tune base models from the Mistral, OLMo, and Phi families on existing public training datasets.
Our replication achieves performance on par with or surpassing existing table LLMs, establishing new state-of-the-art performance on Hitab, a table question-answering dataset.
More importantly, through systematic out-of-domain evaluation, we decouple the contributions of training data and the base model, providing insight into their individual impacts.
In addition, we assess the effects of table-specific instruction tuning on general-purpose benchmarks, revealing trade-offs between specialization and generalization.

Paper Type: Long

Research Area: Special Theme (conference specific)

Research Area Keywords: table instruction tuning, table LLMs, generalization, replication

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Reproduction study, Publicly available software and/or pre-trained models

Languages Studied: English

Submission Number: 143