Language Models as AI Research World Models

2026-10-08

AI Summary

This study addresses the limited capacity of existing AI research agents to predict experimental outcomes, which constrains self-improvement under finite budgets. We propose the Research World Model (RWM), which leverages real experimental data to forecast intervention outcomes and optimize experiment selection strategies. A core contribution is demonstrating that reusing research knowledge across environments significantly enhances prediction accuracy, outperforming approaches that merely scale model size or inference compute. Experiments employing a large language model-based multi-turn autonomous search framework show that our approach reduces environment selection regret by 78% and improves optimal gain by 15.8% for Qwen3. These results validate the critical importance of accumulating experimental data for effective RWM training.

Abstract

AI research agents automate the cycle of proposing, implementing, and evaluating experiments, opening a path toward recursive self-improvement. Yet their ability to propose experiments outpaces their capacity to execute them in real environments, making outcome prediction a key capability for sustained self-improvement under limited experimental budgets. We investigate language models as Research World Models (RWMs), which predict the outcomes of candidate interventions across research environments. Our evaluation draws on over 2,600 experimental records from nine research environments spanning pretraining, post-training, and inference, representing more than 171,000 H100 GPU-hours of experimentation. Research knowledge acquired from real experimental experience improves RWM predictions of unseen interventions within the same environment (Spearman +0.27), and can be reused across environments. For example, using only pretraining experience from OLMo3, Marin, and Nanochat, an RWM reduces selection regret in the Qwen3 environment by 78% compared with zero-experience setting. These benefits extend to multi-round Autoresearch under a fixed selection budget: RWMs with in-env and cross-env research knowledge increase the best gain achieved by 15.8% and 11.6%, respectively. Ablations across 13 language models used as RWMs show that adding research knowledge can improve intervention ranking more than changing models or increasing reasoning effort alone. These findings support language models as RWMs and motivate accumulating experimental data for future RWM training.

Problem

• Research World Models
• outcome prediction
• recursive self-improvement
• AI research agents
• experimental budget

Innovation

• Research World Models
• Language Models
• Autoresearch
• Cross-environment Transfer
• Outcome Prediction