# MLToolBench: Learning Tool-Augmented Agents for Machine Learning Development

arXiv: https://arxiv.org/abs/2609.36679 · HTML: https://mlsanity.com/paper/2609.36679/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 37 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 30 of 32)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [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)
- [Understanding R1-Zero-Like Training: A Critical Perspective](https://mlsanity.com/paper/2503.20783/) (2025)
- [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](https://mlsanity.com/paper/2401.14196/) (2024)
- [API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs](https://mlsanity.com/paper/2304.08244/) (2023)
- [On Variational Bounds of Mutual Information](https://mlsanity.com/paper/1905.06922/) (2019)
- [MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering](https://mlsanity.com/paper/2410.07095/) (2024)
- [ReTool: Reinforcement Learning for Strategic Tool Use in LLMs](https://mlsanity.com/paper/2504.11536/) (2025)
- [MLAgentBench: Evaluating Language Agents on Machine Learning Experimentation](https://mlsanity.com/paper/2310.03302/) (2023)
- [ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities](https://mlsanity.com/paper/2408.04682/) (2024)
- [Unifying distillation and privileged information](https://mlsanity.com/paper/1511.03643/) (2015)
- [AIDE: AI-Driven Exploration in the Space of Code](https://mlsanity.com/paper/2502.13138/) (2025)
- [MLGym: A New Framework and Benchmark for Advancing AI Research Agents](https://mlsanity.com/paper/2502.14499/) (2025)
- [AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench](https://mlsanity.com/paper/2507.02554/) (2025)
- [Hindsight Credit Assignment for Long-Horizon LLM Agents](https://mlsanity.com/paper/2603.08754/) (2026)
- [MLE-Dojo: Interactive Environments for Empowering LLM Agents in Machine Learning Engineering](https://mlsanity.com/paper/2505.07782/) (2025)
- [SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning](https://mlsanity.com/paper/2607.14777/) (2026)
- [Agent0-VL: Exploring Self-Evolving Agent for Tool-Integrated Vision-Language Reasoning](https://mlsanity.com/paper/2511.19900/) (2025)
- [ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering](https://mlsanity.com/paper/2505.23723/) (2025)
- [TIPS: Turn-Level Information-Potential Reward Shaping for Search-Augmented LLMs](https://mlsanity.com/paper/2603.22293/) (2026)
- [Reinforcement Learning for Machine Learning Engineering Agents](https://mlsanity.com/paper/2509.01684/) (2025)
- [Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization](https://mlsanity.com/paper/2605.05040/) (2026)
- [MLE-Smith: Scaling MLE Tasks with Automated Multi-Agent Pipeline](https://mlsanity.com/paper/2510.07307/) (2025)
- [TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning](https://mlsanity.com/paper/2606.32017/) (2026)
- [TACO: Tool-Augmented Credit Optimization for Agentic Tool Use](https://mlsanity.com/paper/2606.30251/) (2026)
- [DeltaML-Bench: Evaluating Machine Learning Agents on Real-World Research Repositories](https://mlsanity.com/paper/2608.19653/) (2026)
- [ML-AutoResearch: Training Machine Learning Research Agents with Automatically Generated Environments](https://mlsanity.com/paper/2603.17216/) (2026)
- [When Should Service Agents Reconsider? Difficulty-Routed Control in Customer-Service Operations](https://mlsanity.com/paper/2607.01426/) (2026)

## Built on by (most cited first, 1 of 1)

- [Verify Less, Evolve More: Training Idea-Level Critics for Verification-Efficient ML Evolving Agents](https://mlsanity.com/paper/2610.08993/) (2026)

## Abstract (arXiv)

Machine learning engineering (MLE) agents have made substantial progress, but learning through ML experimentation remains costly in time and computation. Synthetic environments reduce these costs while introducing variations in data and experimental settings that require task-specific diagnosis. Access to diagnostic tools alone does not ensure that agents learn when to use them or how to act on their findings. We introduce ToolMLBench, a suite of executable tools for data inspection, code verification, and experiment diagnosis, together with an SFT and RL pipeline for learning their use. Diagnostic calls acquire evidence whose value depends on subsequent decisions, so final outcomes provide limited guidance on which calls to reinforce. We address this challenge with SPICE, which measures how privileged context changes the likelihood of a sampled tool action and uses this difference as a turn-level reward alongside the final outcome. We train on 80 synthetic tasks and evaluate on 25 in-domain and 10 out-of-domain tasks. Providing tool interfaces and descriptions alone yields inconsistent gains across unadapted models. With the same diagnostic interface, our training pipeline raises in-domain success from 24.8% to 52.4% for Qwen3-8B and from 35.6% to 69.2% for Qwen3.5-35B-A3B. The latter also improves from 31% to 48% out-of-domain, supporting learned diagnostic tool use on held-out sources and targets.