Related papers: LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities

LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities

URL: http://arxiv.org/abs/2504.16078v1
Date: Tue, 22 Apr 2025 17:57:14 GMT
Title: LLMs are Greedy Agents: Effects of RL Fine-tuning on Decision-Making Abilities
Authors: Thomas Schmied, Jörg Bornschein, Jordi Grau-Moya, Markus Wulfmeier, Razvan Pascanu,
Abstract summary: We study why Large Language Models (LLMs) perform sub-optimally in decision-making scenarios.<n>We propose mitigation of these shortcomings by fine-tuning via Reinforcement Learning (RL) on self-generated CoT rationales.
Score: 21.42711537107199
License: http://creativecommons.org/licenses/by/4.0/
Abstract: The success of Large Language Models (LLMs) has sparked interest in various agentic applications. A key hypothesis is that LLMs, leveraging common sense and Chain-of-Thought (CoT) reasoning, can effectively explore and efficiently solve complex domains. However, LLM agents have been found to suffer from sub-optimal exploration and the knowing-doing gap, the inability to effectively act on knowledge present in the model. In this work, we systematically study why LLMs perform sub-optimally in decision-making scenarios. In particular, we closely examine three prevalent failure modes: greediness, frequency bias, and the knowing-doing gap. We propose mitigation of these shortcomings by fine-tuning via Reinforcement Learning (RL) on self-generated CoT rationales. Our experiments across multi-armed bandits, contextual bandits, and Tic-tac-toe, demonstrate that RL fine-tuning enhances the decision-making abilities of LLMs by increasing exploration and narrowing the knowing-doing gap. Finally, we study both classic exploration mechanisms, such as $\epsilon$-greedy, and LLM-specific approaches, such as self-correction and self-consistency, to enable more effective fine-tuning of LLMs for decision-making.

Related papers

R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning [87.30285670315334]
textbfR1-Searcher is a novel two-stage outcome-based RL approach designed to enhance the search capabilities of Large Language Models.<n>Our framework relies exclusively on RL, without requiring process rewards or distillation for a cold start.<n>Our experiments demonstrate that our method significantly outperforms previous strong RAG methods, even when compared to the closed-source GPT-4o-mini.
arXiv Detail & Related papers (2025-03-07T17:14:44Z)
Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search [57.28671084993782]
Large language models (LLMs) have demonstrated remarkable reasoning capabilities across diverse domains.<n>Recent studies have shown that increasing test-time computation enhances LLMs' reasoning capabilities.<n>We propose a two-stage training paradigm: 1) a small-scale format tuning stage to internalize the COAT reasoning format and 2) a large-scale self-improvement stage leveraging reinforcement learning.
arXiv Detail & Related papers (2025-02-04T17:26:58Z)
Large Language Models Think Too Fast To Explore Effectively [0.0]
The extent to which Large Language Models can effectively explore, particularly in open-ended tasks, remains unclear.<n>This study investigates whether LLMs can surpass humans in exploration during an open-ended task, using Little Alchemy 2 as a paradigm.
arXiv Detail & Related papers (2025-01-29T21:51:17Z)
Efficient Reinforcement Learning with Large Language Model Priors [18.72288751305885]
Large language models (LLMs) have recently emerged as powerful general-purpose tools. We propose treating LLMs as prior action distributions and integrating them into RL frameworks. We show that incorporating LLM-based action priors significantly reduces exploration and complexity optimization.
arXiv Detail & Related papers (2024-10-10T13:54:11Z)
EVOLvE: Evaluating and Optimizing LLMs For Exploration [76.66831821738927]
Large language models (LLMs) remain under-studied in scenarios requiring optimal decision-making under uncertainty. We measure LLMs' (in)ability to make optimal decisions in bandits, a state-less reinforcement learning setting relevant to many applications. Motivated by the existence of optimal exploration algorithms, we propose efficient ways to integrate this algorithmic knowledge into LLMs.
arXiv Detail & Related papers (2024-10-08T17:54:03Z)
Towards Efficient LLM Grounding for Embodied Multi-Agent Collaboration [70.09561665520043]
We propose a novel framework for multi-agent collaboration that introduces Reinforced Advantage feedback (ReAd) for efficient self-refinement of plans. We provide theoretical analysis by extending advantage-weighted regression in reinforcement learning to multi-agent systems. Experiments on Over-AI and a difficult variant of RoCoBench show that ReAd surpasses baselines in success rate, and also significantly decreases the interaction steps of agents.
arXiv Detail & Related papers (2024-05-23T08:33:19Z)
Small Models, Big Insights: Leveraging Slim Proxy Models To Decide When and What to Retrieve for LLMs [60.40396361115776]
This paper introduces a novel collaborative approach, namely SlimPLM, that detects missing knowledge in large language models (LLMs) with a slim proxy model. We employ a proxy model which has far fewer parameters, and take its answers as answers. Heuristic answers are then utilized to predict the knowledge required to answer the user question, as well as the known and unknown knowledge within the LLM.
arXiv Detail & Related papers (2024-02-19T11:11:08Z)
Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificity [61.54815512469125]
This survey addresses the crucial issue of factuality in Large Language Models (LLMs) As LLMs find applications across diverse domains, the reliability and accuracy of their outputs become vital.
arXiv Detail & Related papers (2023-10-11T14:18:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.