# EdgeLM: Edge Demonstrations for Language Models' Table Understanding

arXiv: https://arxiv.org/abs/2608.04390 · HTML: https://mlsanity.com/paper/2608.04390/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 49 references that are in the map.
- First posted on arXiv 5 Aug 2026.

## Builds on (most cited first, 28 of 28)

- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?](https://mlsanity.com/paper/2202.12837/) (2022)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [What Makes Good In-Context Examples for GPT-3?](https://mlsanity.com/paper/2101.06804/) (2021)
- [Learning To Retrieve Prompts for In-Context Learning](https://mlsanity.com/paper/2112.08633/) (2021)
- [OpenML: networked science in machine learning](https://mlsanity.com/paper/1407.7722/) (2014)
- [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://mlsanity.com/paper/2004.02349/) (2020)
- [TaBERT: Pretraining for Joint Understanding of Textual and Tabular Data](https://mlsanity.com/paper/2005.08314/) (2020)
- [Language Models are Realistic Tabular Data Generators](https://mlsanity.com/paper/2210.06280/) (2022)
- [Selective Annotation Makes Language Models Better Few-Shot Learners](https://mlsanity.com/paper/2209.01975/) (2022)
- [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://mlsanity.com/paper/2107.07653/) (2021)
- [Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding](https://mlsanity.com/paper/2401.04398/) (2024)
- [Can Foundation Models Wrangle Your Data?](https://mlsanity.com/paper/2205.09911/) (2022)
- [Large Language Models(LLMs) on Tabular Data: Prediction, Generation, and Understanding -- A Survey](https://mlsanity.com/paper/2402.17944/) (2024)
- [Deep Entity Matching with Pre-Trained Language Models](https://mlsanity.com/paper/2004.00584/) (2020)
- [Diverse Demonstrations Improve In-context Compositional Generalization](https://mlsanity.com/paper/2212.06800/) (2022)
- [TableLlama: Towards Open Large Generalist Models for Tables](https://mlsanity.com/paper/2311.09206/) (2023)
- [TableFormer: Robust Transformer Modeling for Table-Text Encoding](https://mlsanity.com/paper/2203.00274/) (2022)
- [In-Context Learning with Iterative Demonstration Selection](https://mlsanity.com/paper/2310.09881/) (2023)
- [TableRAG: Million-Token Table Understanding with Language Models](https://mlsanity.com/paper/2410.04739/) (2024)
- [Valentine: Evaluating Matching Techniques for Dataset Discovery](https://mlsanity.com/paper/2010.07386/) (2020)
- [LLM-Forest: Ensemble Learning of LLMs with Graph-Augmented Prompts for Data Imputation](https://mlsanity.com/paper/2410.21520/) (2024)
- [Anomaly Detection of Tabular Data Using LLMs](https://mlsanity.com/paper/2406.16308/) (2024)
- [On LLM-Enhanced Mixed-Type Data Imputation with High-Order Message Passing](https://mlsanity.com/paper/2501.02191/) (2025)
- [Truth Finding on the Deep Web: Is the Problem Solved?](https://mlsanity.com/paper/1503.00303/) (2015)
- [LDI: Localized Data Imputation for Text-Rich Tables](https://mlsanity.com/paper/2506.16616/) (2025)
- [SynSQL: Synthesizing Relational Databases for Robust Evaluation of Text-to-SQL Systems](https://mlsanity.com/paper/2604.27261/) (2026)
- [SQL-Exchange: Transforming SQL Queries Across Domains](https://mlsanity.com/paper/2508.07087/) (2025)

## Abstract (arXiv)

Large language models (LLMs) perform table-centric prediction through in-context learning, making demonstration selection critical to performance. Existing retrieval methods prioritize similarity to the query, but similar demonstrations often reinforce the model's likely prediction rather than reveal the distinctions needed for difficult decisions. We propose EdgeLM, a retrieval framework that instead selects edge evidence, demonstrations that are both relevant to the query and informative about the decision boundary. EdgeLM retrieves two complementary forms of edge evidence by selecting data edges, nearby examples with different ground-truth labels, and model edges, similar examples previously misclassified by the deployed model. EdgeLM requires neither model retraining nor task-specific engineering. Across five data wrangling tasks, fifteen datasets, and five open-weight and proprietary LLMs, EdgeLM consistently achieves the best or near-best performance in every setting, while ablations show that the two forms of edge evidence provide complementary benefits. Our code and datasets are publicly available at https://github.com/soroushomidvar/EdgeLM.