# COSMosFL: Ensemble of Small Language Models for Fault Localisation

arXiv: https://arxiv.org/abs/2502.02908 · HTML: https://mlsanity.com/paper/2502.02908/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 44 references that are in the map.
- First posted on arXiv 5 Feb 2025.

## Builds on (most cited first, 23 of 23)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [RouteLLM: Learning to Route LLMs with Preference Data](https://mlsanity.com/paper/2406.18665/) (2024)
- [Mixture-of-Experts with Expert Choice Routing](https://mlsanity.com/paper/2202.09368/) (2022)
- [LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion](https://mlsanity.com/paper/2306.02561/) (2023)
- [RepairAgent: An Autonomous, LLM-Based Agent for Program Repair](https://mlsanity.com/paper/2403.17134/) (2024)
- [Large Language Models for Test-Free Fault Localization](https://mlsanity.com/paper/2310.01726/) (2023)
- [Learning to Decode Collaboratively with Multiple Language Models](https://mlsanity.com/paper/2403.03870/) (2024)
- [A Quantitative and Qualitative Evaluation of LLM-Based Explainable Fault Localization](https://mlsanity.com/paper/2308.05487/) (2023)
- [Better patching using LLM prompting, via Self-Consistency](https://mlsanity.com/paper/2306.00108/) (2023)
- [Large Language Models in Fault Localisation](https://mlsanity.com/paper/2308.15276/) (2023)
- [Arachne: Search Based Repair of Deep Neural Networks](https://mlsanity.com/paper/1912.12463/) (2019)
- [Towards Autonomous Testing Agents via Conversational Large Language Models](https://mlsanity.com/paper/2306.05152/) (2023)
- [Evaluating Diverse Large Language Models for Automatic and General Bug Reproduction](https://mlsanity.com/paper/2311.04532/) (2023)
- [InfeRE: Step-by-Step Regex Generation via Chain of Inference](https://mlsanity.com/paper/2308.04041/) (2023)

## Built on by (most cited first, 3 of 3)

- [The Rise of Small Language Models in Healthcare: A Comprehensive Survey](https://mlsanity.com/paper/2504.17119/) (2025)
- [Atropos: Improving Cost-Benefit Trade-off of LLM-based Agents under Self-Consistency with Early Termination and Model Hotswap](https://mlsanity.com/paper/2604.15075/) (2026)
- [SLMJury: Can Small Language Models Judge as Well as Large Ones?](https://mlsanity.com/paper/2606.07810/) (2026)

## Abstract (arXiv)

LLMs are rapidly being adopted to build powerful tools and agents for software engineering, but most of them rely heavily on extremely large closed-source models. This, in turn, can hinder wider adoption due to security issues as well as financial cost and environmental impact. Recently, a number of open source Small Language Models (SLMs) are being released and gaining traction. While SLMs are smaller, more energy-efficient, and therefore easier to locally deploy, they tend to show worse performance when compared to larger closed LLMs. We present COSMos, a task-level LLM ensemble technique that uses voting mechanism, to provide a broader range of choice between SLMs and LLMs. We instantiate COSMos with an LLM-based Fault Localisation technique, AutoFL, and report the cost-benefit trade-off between LLM accuracy and various costs such as energy consumption, inference time, and the number of tokens used. An empirical evaluation using Defects4J shows that COSMos can build effective ensembles that can achieve Pareto-optimality in terms of FL accuracy and inference cost, when compared to individual models.