# Magnet: Multi-turn Tool-use Data Synthesis and Distillation via Graph Translation

arXiv: https://arxiv.org/abs/2503.07826 · HTML: https://mlsanity.com/paper/2503.07826/

## Key facts

- Cited by 37 papers in the map (data through 14 Oct 2026).
- Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use.
- Builds on 28 of its 32 references that are in the map.
- First posted on arXiv 10 Mar 2025.

## Placed in

- LLM agents and tool use / [Tool learning and function calling](https://mlsanity.com/agents/tool-learning-and-function-calling/)
- Synthetic data and self-improvement for LLMs / [LLM synthetic data generation](https://mlsanity.com/synth/llm-synthetic-data-generation/)

## Builds on (most cited first, 28 of 28)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [Gorilla: Large Language Model Connected with Massive APIs](https://mlsanity.com/paper/2305.15334/) (2023)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents](https://mlsanity.com/paper/2401.13178/) (2024)
- [ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities](https://mlsanity.com/paper/2408.04682/) (2024)
- [ToolACE: Winning the Points of LLM Function Calling](https://mlsanity.com/paper/2409.00920/) (2024)
- [APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasets](https://mlsanity.com/paper/2406.18518/) (2024)
- [Self-Alignment with Instruction Backtranslation](https://mlsanity.com/paper/2308.06259/) (2023)
- [Learning by Distilling Context](https://mlsanity.com/paper/2209.15189/) (2022)
- [StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models](https://mlsanity.com/paper/2403.07714/) (2024)
- [xLAM: A Family of Large Action Models to Empower AI Agent Systems](https://mlsanity.com/paper/2409.03215/) (2024)
- [Agent Lumos: Unified and Modular Training for Open-Source Language Agents](https://mlsanity.com/paper/2311.05657/) (2023)
- [Building Math Agents with Multi-Turn Iterative Preference Learning](https://mlsanity.com/paper/2409.02392/) (2024)
- [Hammer: Robust Function-Calling for On-Device Language Models via Function Masking](https://mlsanity.com/paper/2410.04587/) (2024)
- [Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks](https://mlsanity.com/paper/2407.00121/) (2024)
- [Re-Invoke: Tool Invocation Rewriting for Zero-Shot Tool Retrieval](https://mlsanity.com/paper/2408.01875/) (2024)
- [NESTFUL: A Benchmark for Evaluating LLMs on Nested Sequences of API Calls](https://mlsanity.com/paper/2409.03797/) (2024)
- [Facilitating Multi-turn Function Calling for LLMs via Compositional Instruction Tuning](https://mlsanity.com/paper/2410.12952/) (2024)
- [Better Alignment with Instruction Back-and-Forth Translation](https://mlsanity.com/paper/2408.04614/) (2024)

## Built on by (most cited first, 30 of 37)

- [APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplay](https://mlsanity.com/paper/2504.03601/) (2025)
- [A Survey of Context Engineering for Large Language Models](https://mlsanity.com/paper/2507.13334/) (2025)
- [Nemotron-Research-Tool-N1: Exploring Tool-Using Language Models with Reinforced Reasoning](https://mlsanity.com/paper/2505.00024/) (2025)
- [Towards General Agentic Intelligence via Environment Scaling](https://mlsanity.com/paper/2509.13311/) (2025)
- [Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents](https://mlsanity.com/paper/2602.13379/) (2026)
- [Don't Just Fine-tune the Agent, Tune the Environment](https://mlsanity.com/paper/2510.10197/) (2025)
- [LongFuncEval: Measuring the effectiveness of long context models for function calling](https://mlsanity.com/paper/2505.10570/) (2025)
- [LoopTool: Closing the Data-Training Loop for Robust LLM Tool Calls](https://mlsanity.com/paper/2511.09148/) (2025)
- [EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL](https://mlsanity.com/paper/2605.18703/) (2026)
- [Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text](https://mlsanity.com/paper/2601.10355/) (2026)
- [GTool: Graph Enhanced Tool Planning with Large Language Model](https://mlsanity.com/paper/2508.12725/) (2025)
- [TURA: Tool-Augmented Unified Retrieval Agent for AI Search](https://mlsanity.com/paper/2508.04604/) (2025)
- [Experience-Evolving Multi-Turn Tool-Use Agent with Hybrid Episodic-Procedural Memory](https://mlsanity.com/paper/2512.07287/) (2025)
- [AgentSkiller: Scaling Generalist Agent Intelligence through Semantically Integrated Cross-Domain Data Synthesis](https://mlsanity.com/paper/2602.09372/) (2026)
- [DRAGON: Domain-specific Robust Automatic Data Generation for RAG Optimization](https://mlsanity.com/paper/2505.10989/) (2025)
- [PARL-MT: Learning to Call Functions in Multi-Turn Conversation with Progress Awareness](https://mlsanity.com/paper/2509.23206/) (2025)
- [How Can Input Reformulation Improve Tool Usage Accuracy in a Complex Dynamic Environment? A Study on τ-bench](https://mlsanity.com/paper/2508.20931/) (2025)
- [Simulating Complex Multi-Turn Tool Calling Interactions in Stateless Execution Environments](https://mlsanity.com/paper/2601.19914/) (2026)
- [UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents](https://mlsanity.com/paper/2604.11557/) (2026)
- [What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agents](https://mlsanity.com/paper/2608.27260/) (2026)
- [MagicAgent: Towards Generalized Agent Planning](https://mlsanity.com/paper/2602.19000/) (2026)
- [RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents](https://mlsanity.com/paper/2606.19047/) (2026)
- [GOAT: A Training Framework for Goal-Oriented Agent with Tools](https://mlsanity.com/paper/2510.12218/) (2025)
- [Linguistic and Argument Diversity in Synthetic Data for Function-Calling Agents](https://mlsanity.com/paper/2601.17829/) (2026)
- [Beyond Accuracy: A Cognitive Load Framework for Mapping the Capability Boundaries of Tool-use Agents](https://mlsanity.com/paper/2601.20412/) (2026)
- [ToolWeave: Structured Synthesis of Complex Multi-Turn Tool-Calling Dialogues](https://mlsanity.com/paper/2605.12521/) (2026)
- [Scaling Agentic Capabilities via Grounded Interaction Synthesis](https://mlsanity.com/paper/2606.02001/) (2026)
- [Synthesize and Reward -- Reinforcement Learning for Multi-Step Tool Use in Live Environments](https://mlsanity.com/paper/2606.03892/) (2026)
- [WeClawArena: An Auditable Sandbox and Benchmark for Cross-User Agents Collaboration and Security in Human-Centered Agent Networks](https://mlsanity.com/paper/2608.03499/) (2026)
- [Towards Better Agents for Multi-Turn User Interaction: The Next User Turn Is More Than Context](https://mlsanity.com/paper/2608.17499/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have exhibited the ability to effectively utilize external tools to address user queries. However, their performance may be limited in complex, multi-turn interactions involving users and multiple tools. To address this, we propose Magnet, a principled framework for synthesizing high-quality training trajectories to enhance the function calling capability of large language model agents in multi-turn conversations with humans. The framework is based on automatic and iterative translations from a function signature path to a sequence of queries and executable function calls. We model the complicated function interactions in multi-turn cases with graph and design novel node operations to build reliable signature paths. Motivated by context distillation, when guiding the generation of positive and negative trajectories using a teacher model, we provide reference function call sequences as positive hints in context and contrastive, incorrect function calls as negative hints. Experiments show that training with the positive trajectories with supervised fine-tuning and preference optimization against negative trajectories, our 14B model, Magnet-14B-mDPO, obtains 68.01 on BFCL-v3 and 73.30 on ToolQuery, surpassing the performance of the teacher model Gemini-1.5-pro-002 by a large margin in function calling.