EXPLORA: Efficient Exemplar Subset Selection for Complex Reasoning

Kiran Purohit, Venktesh V, Raghuram Devalla, Krishna Mohan Yerragorla, Sourangshu Bhattacharya, Avishek Anand


Abstract
Answering reasoning-based complex questions over text and hybrid sources, including tables, is a challenging task. Recent advances in large language models (LLMs) have enabled in-context learning (ICL), allowing LLMs to acquire proficiency in a specific task using only a few demonstration samples (exemplars). A critical challenge in ICL is the selection of optimal exemplars, which can be either task-specific (static) or test-example-specific (dynamic). Static exemplars provide faster inference times and increased robustness across a distribution of test examples. In this paper, we propose an algorithm for static exemplar subset selection for complex reasoning tasks. We introduce EXPLORA, a novel exploration method designed to estimate the parameters of the scoring function, which evaluates exemplar subsets without incorporating confidence information. EXPLORA significantly reduces the number of LLM calls to ~11% of those required by state-of-the-art methods and achieves a substantial performance improvement of 12.24%. We open-source our code and data (https://github.com/kiranpurohit/EXPLORA).
Anthology ID:
2024.emnlp-main.307
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5367–5388
Language:
URL:
https://aclanthology.org/2024.emnlp-main.307/
DOI:
10.18653/v1/2024.emnlp-main.307
Bibkey:
Cite (ACL):
Kiran Purohit, Venktesh V, Raghuram Devalla, Krishna Mohan Yerragorla, Sourangshu Bhattacharya, and Avishek Anand. 2024. EXPLORA: Efficient Exemplar Subset Selection for Complex Reasoning. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 5367–5388, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
EXPLORA: Efficient Exemplar Subset Selection for Complex Reasoning (Purohit et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.307.pdf
Software:
 2024.emnlp-main.307.software.zip
Data:
 2024.emnlp-main.307.data.zip