# Search-based Optimisation of LLM Learning Shots for Story Point Estimation

arXiv: https://arxiv.org/abs/2403.08430 · HTML: https://mlsanity.com/paper/2403.08430/

## Key facts

- Cited by 5 papers in the map (data through 7 Oct 2026).
- Builds on 4 of its 10 references that are in the map.
- First posted on arXiv 13 Mar 2024.

## Builds on (most cited first, 4 of 4)

- [FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance](https://mlsanity.com/paper/2305.05176/) (2023)
- [An Empirical Study of the Non-determinism of ChatGPT in Code Generation](https://mlsanity.com/paper/2308.02828/) (2023)
- [A Versatile Dataset of Agile Open Source Software Projects](https://mlsanity.com/paper/2202.00979/) (2022)
- [Agile Effort Estimation: Have We Solved the Problem Yet? Insights From A Replication Study](https://mlsanity.com/paper/2201.05401/) (2022)

## Built on by (most cited first, 5 of 5)

- [Search-Based LLMs for Code Optimization](https://mlsanity.com/paper/2408.12159/) (2024)
- [Meta-RAG on Large Codebases Using Code Summarization](https://mlsanity.com/paper/2508.02611/) (2025)
- [ALMAS: an Autonomous LLM-based Multi-Agent Software Engineering Framework](https://mlsanity.com/paper/2510.03463/) (2025)
- [Search-Based Software Engineering and AI Foundation Models: Current Landscape and Future Roadmap](https://mlsanity.com/paper/2505.19625/) (2025)
- [Agile Story-Point Estimation: Is RAG a Better Way to Go?](https://mlsanity.com/paper/2604.03443/) (2026)

## Abstract (arXiv)

One of the ways Large Language Models (LLMs) are used to perform machine learning tasks is to provide them with a few examples before asking them to produce a prediction. This is a meta-learning process known as few-shot learning. In this paper, we use available Search-Based methods to optimise the number and combination of examples that can improve an LLM's estimation performance, when it is used to estimate story points for new agile tasks. Our preliminary results show that our SBSE technique improves the estimation performance of the LLM by 59.34% on average (in terms of mean absolute error of the estimation) over three datasets against a zero-shot setting.