# Meta-Reasoning Improves Tool Use in Large Language Models

arXiv: https://arxiv.org/abs/2411.04535 · HTML: https://mlsanity.com/paper/2411.04535/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 16 of its 41 references that are in the map.
- First posted on arXiv 7 Nov 2024.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)
- LLM agents and tool use / [Tool learning and function calling](https://mlsanity.com/agents/tool-learning-and-function-calling/)

## Builds on (most cited first, 16 of 16)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Gorilla: Large Language Model Connected with Massive APIs](https://mlsanity.com/paper/2305.15334/) (2023)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases](https://mlsanity.com/paper/2306.05301/) (2023)
- [ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving](https://mlsanity.com/paper/2309.17452/) (2023)
- [Think before you speak: Training Language Models With Pause Tokens](https://mlsanity.com/paper/2310.02226/) (2023)
- [TALM: Tool Augmented Language Models](https://mlsanity.com/paper/2205.12255/) (2022)
- [ART: Automatic multi-step reasoning and tool-use for large language models](https://mlsanity.com/paper/2303.09014/) (2023)
- [DE-COP: Detecting Copyrighted Content in Language Models Training Data](https://mlsanity.com/paper/2402.09910/) (2024)
- [Tool Documentation Enables Zero-Shot Tool-Usage with Large Language Models](https://mlsanity.com/paper/2308.00675/) (2023)
- [MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation](https://mlsanity.com/paper/2312.17080/) (2023)
- [Thinking Tokens for Language Modeling](https://mlsanity.com/paper/2405.08644/) (2024)

## Built on by (most cited first, 6 of 6)

- [ToolScope: Enhancing LLM Agent Tool Use through Tool Merging and Context-Aware Filtering](https://mlsanity.com/paper/2510.20036/) (2025)
- [JTPRO: A Joint Tool-Prompt Reflective Optimization Framework for Language Agents](https://mlsanity.com/paper/2604.19821/) (2026)
- [Task-Aligned Tool Recommendation for Large Language Models](https://mlsanity.com/paper/2411.09613/) (2024)
- [AgentCoMa: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World Scenarios](https://mlsanity.com/paper/2508.19988/) (2025)
- [Searching Meta Reasoning Skeleton to Guide LLM Reasoning](https://mlsanity.com/paper/2510.04116/) (2025)
- [Meta-Moderator: Empowering Multi-Agent Debate with Meta-Cognition](https://mlsanity.com/paper/2608.23029/) (2026)

## Abstract (arXiv)

External tools help large language models succeed at tasks where they would otherwise typically fail. In existing frameworks, choosing tools at test time relies on naive greedy decoding, regardless of whether the model has been fine-tuned on tool-annotated data or prompted with in-context examples. In contrast, we find that gathering and choosing among a suitable set of candidate tools has greater potential to lead to an optimal selection. We present Tool selECTion via meta-reasONing (TECTON), a two-phase system that first reasons over a task and outputs candidate tools using a custom fine-tuned language modelling head. Then, with the custom head disabled, it meta-reasons (i.e., it reasons over the previous reasoning process) to make a final choice. We show that TECTON results in substantial gains--both in-distribution and out-of-distribution--on a range of math reasoning datasets.