# Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search

arXiv: https://arxiv.org/abs/2603.01692 · HTML: https://mlsanity.com/paper/2603.01692/

## Key facts

- Cited by 4 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 20 of its 35 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## 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 / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 20 of 20)

- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [Automatic Prompt Optimization with "Gradient Descent" and Beam Search](https://mlsanity.com/paper/2305.03495/) (2023)
- [MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering](https://mlsanity.com/paper/2410.07095/) (2024)
- [MLAgentBench: Evaluating Language Agents on Machine Learning Experimentation](https://mlsanity.com/paper/2310.03302/) (2023)
- [Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models](https://mlsanity.com/paper/2501.09686/) (2025)
- [TextGrad: Automatic "Differentiation" via Text](https://mlsanity.com/paper/2406.07496/) (2024)
- [AIDE: AI-Driven Exploration in the Space of Code](https://mlsanity.com/paper/2502.13138/) (2025)
- [AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench](https://mlsanity.com/paper/2507.02554/) (2025)
- [Process vs. Outcome Reward: Which is Better for Agentic RAG Reinforcement Learning](https://mlsanity.com/paper/2505.14069/) (2025)
- [ML-Master: Towards AI-for-AI via Integration of Exploration and Reasoning](https://mlsanity.com/paper/2506.16499/) (2025)
- [MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinement](https://mlsanity.com/paper/2506.15692/) (2025)
- [The FM Agent](https://mlsanity.com/paper/2510.26144/) (2025)
- [DSGym: A Holistic Framework for Evaluating and Training Data Science Agents](https://mlsanity.com/paper/2601.16344/) (2026)
- [Introducing MAPO: Momentum-Aided Gradient Descent Prompt Optimization](https://mlsanity.com/paper/2410.19499/) (2024)
- [KompeteAI: Accelerated Autonomous Multi-Agent System for End-to-End Pipeline Generation for Machine Learning Problems](https://mlsanity.com/paper/2508.10177/) (2025)
- [Comparative Analysis of Gradient-Based Optimization Techniques Using Multidimensional Surface 3D Visualizations and Initial Point Sensitivity](https://mlsanity.com/paper/2409.04470/) (2024)

## Built on by (most cited first, 4 of 4)

- [MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery](https://mlsanity.com/paper/2606.06473/) (2026)
- [RecHarness: A Bandit-Routed Agentic Harness for Self-Evolving Recommender Systems](https://mlsanity.com/paper/2607.29241/) (2026)
- [Repo-To-Skill: Distilling GitHub Repositories Into AI4AI Skills](https://mlsanity.com/paper/2609.02749/) (2026)
- [How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?](https://mlsanity.com/paper/2609.40303/) (2026)

## Abstract (arXiv)

LLM-based agents for machine learning engineering (MLE) predominantly rely on tree search, a form of gradient-free optimization that uses scalar validation scores to rank candidates. As LLM reasoning capabilities improve, exhaustive enumeration becomes increasingly inefficient compared to directed updates, analogous to how accurate gradients enable efficient descent over random search. We introduce Gome, an MLE agent that operationalizes gradient-based optimization. Gome maps structured diagnostic reasoning to gradient computation, success memory to momentum, and multi-trace execution to distributed optimization. Under a closed-world protocol that isolates architectural effects from external knowledge, Gome achieves a state-of-the-art 35.1% any-medal rate on MLE-Bench with a restricted 12-hour budget on a single V100 GPU. Scaling experiments across 10 models reveal a critical crossover: with weaker models, tree search retains advantages by compensating for unreliable reasoning through exhaustive exploration; as reasoning capability strengthens, gradient-based optimization progressively outperforms, with the gap widening at frontier-tier models. Given the rapid advancement of reasoning-oriented LLMs, this positions gradient-based optimization as an increasingly favorable paradigm. We release our codebase and GPT-5 traces at https://github.com/microsoft/RD-Agent.