# Data Efficient Sample Selection for In-Context Learning

arXiv: https://arxiv.org/abs/2609.06670 · HTML: https://mlsanity.com/paper/2609.06670/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 34 of its 47 references that are in the map.
- First posted on arXiv 6 Sep 2026.

## Builds on (most cited first, 30 of 34)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps](https://mlsanity.com/paper/2011.01060/) (2020)
- [Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks](https://mlsanity.com/paper/2211.12588/) (2022)
- [Learning To Retrieve Prompts for In-Context Learning](https://mlsanity.com/paper/2112.08633/) (2021)
- [Train faster, generalize better: Stability of stochastic gradient descent](https://mlsanity.com/paper/1509.01240/) (2015)
- [Complexity-Based Prompting for Multi-Step Reasoning](https://mlsanity.com/paper/2210.00720/) (2022)
- [Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning](https://mlsanity.com/paper/2209.14610/) (2022)
- [Trained Transformers Learn Linear Models In-Context](https://mlsanity.com/paper/2306.09927/) (2023)
- [Active Example Selection for In-Context Learning](https://mlsanity.com/paper/2211.04486/) (2022)
- [DeepCore: A Comprehensive Library for Coreset Selection in Deep Learning](https://mlsanity.com/paper/2204.08499/) (2022)
- [Compositional Exemplars for In-context Learning](https://mlsanity.com/paper/2302.05698/) (2023)
- [Unified Demonstration Retriever for In-Context Learning](https://mlsanity.com/paper/2305.04320/) (2023)
- [Finding Support Examples for In-Context Learning](https://mlsanity.com/paper/2302.13539/) (2023)
- [Sequential Experimental Design for Transductive Linear Bandits](https://mlsanity.com/paper/1906.08399/) (2019)
- [Optimal Best-arm Identification in Linear Bandits](https://mlsanity.com/paper/2006.16073/) (2020)
- [Complementary Explanations for Effective In-Context Learning](https://mlsanity.com/paper/2211.13892/) (2022)
- [Gamification of Pure Exploration for Linear Bandits](https://mlsanity.com/paper/2007.00953/) (2020)
- [Fully adaptive algorithm for pure exploration in linear bandits](https://mlsanity.com/paper/1710.05552/) (2017)
- [NeuralNDCG: Direct Optimisation of a Ranking Metric via Differentiable Relaxation of Sorting](https://mlsanity.com/paper/2102.07831/) (2021)
- [Nearly Instance Optimal Sample Complexity Bounds for Top-k Arm Selection](https://mlsanity.com/paper/1702.03605/) (2017)
- [PiRank: Scalable Learning To Rank via Differentiable Sorting](https://mlsanity.com/paper/2012.06731/) (2020)
- [Online Learning to Rank in Stochastic Click Models](https://mlsanity.com/paper/1703.02527/) (2017)
- [Context-Aware Learning to Rank with Self-Attention](https://mlsanity.com/paper/2005.10084/) (2020)
- [Online Learning to Rank with Features](https://mlsanity.com/paper/1810.02567/) (2018)
- [BERTTune: Fine-Tuning Neural Machine Translation with BERTScore](https://mlsanity.com/paper/2106.02208/) (2021)
- [Sample Efficient Demonstration Selection for In-Context Learning](https://mlsanity.com/paper/2506.08607/) (2025)
- [Top-m identification for linear bandits](https://mlsanity.com/paper/2103.10070/) (2021)
- [SEER : A Knapsack approach to Exemplar Selection for In-Context HybridQA](https://mlsanity.com/paper/2310.06675/) (2023)

## Abstract (arXiv)

The In-context learning (ICL) paradigm aids large language models (LLMs) to adapt to new tasks without need for fine-tuning. However, selecting an optimal combination of demonstration examples from a large pool of example subsets is a challenging problem. Existing approaches for selection do not model the complex relationship between ICL samples and downstream LLM performance. They typically perform static task-level selection, choosing subsets once offline, which can fail to generalize to unseen queries. We introduce DearICL (Data Efficient Algorithm for Ranking) ICL samples, a new framework that models demonstration example selection as a subset ranking problem. DearICL employs a non-linear surrogate employing a differentiable sorting objective within a gap-index bandit algorithm. The gap-index based approach enables fine-grained separation of good arms and borderline arms, which is used as an auxiliary objective to train the non-linear surrogate through sufficient sampling of borderline arms, supporting instance-level subset ranking. On exemplar selection benchmarks with open-source LLMs, DearICL achieves 8.08-15.9% accuracy gains over strong linear bandit baselines, with low sample complexity. Code and data: https://github.com/VenkteshV/DearICL.