Cover Image

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

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Schmied, Thomas, Bornschein, Jörg, Grau-Moya, Jordi, Wulfmeier, Markus, Pascanu, Razvan
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Retrieval-Augmented Decision Transformer: External Memory for In-context RL
by: Schmied, Thomas, et al.
Published: (2024)

Fine-Tuned In-Context Learners for Efficient Adaptation
by: Bornschein, Jorg, et al.
Published: (2025)

Kalman Filter for Online Classification of Non-Stationary Data
by: Titsias, Michalis K., et al.
Published: (2023)

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks
by: Schmied, Thomas, et al.
Published: (2024)

Lattice: Learning to Efficiently Compress the Memory
by: Karami, Mahdi, et al.
Published: (2025)

Meta-learning how to Share Credit among Macro-Actions
by: Hosu, Ionel-Alexandru, et al.
Published: (2025)

Revisiting Adam for Streaming Reinforcement Learning
by: Gogianu, Florin, et al.
Published: (2026)

Parameter Efficient Fine-tuning via Explained Variance Adaptation
by: Paischer, Fabian, et al.
Published: (2024)

Superhuman Safe and Agile Racing through Multi-Agent Reinforcement Learning
by: Geles, Ismail, et al.
Published: (2026)

Activated LoRA: Fine-tuned LLMs for Intrinsics
by: Greenewald, Kristjan, et al.
Published: (2025)

Softmax is not Enough (for Sharp Size Generalisation)
by: Veličković, Petar, et al.
Published: (2024)

On the generalization of language models from in-context learning and finetuning: a controlled study
by: Lampinen, Andrew K., et al.
Published: (2025)

Partition Tree Weighting for Non-Stationary Stochastic Bandits
by: Veness, Joel, et al.
Published: (2025)

State Soup: In-Context Skill Learning, Retrieval and Mixing
by: Pióro, Maciej, et al.
Published: (2024)

DARA: Few-shot Budget Allocation in Online Advertising via In-Context Decision Making with RL-Finetuned LLMs
by: Song, Mingxuan, et al.
Published: (2026)

Mining Generalizable Activation Functions
by: Vitvitskyi, Alex, et al.
Published: (2026)

Perplexity Cannot Always Tell Right from Wrong
by: Veličković, Petar, et al.
Published: (2026)

Layerwise LQR for Geometry-Aware Optimization of Deep Networks
by: Dufort-Labbé, Simon, et al.
Published: (2026)

Navigating Potholes with Geometry-Aware Sharpness Minimization
by: Dufort-Labbé, Simon, et al.
Published: (2026)

Growing Q-Networks: Solving Continuous Control Tasks with Adaptive Control Resolution
by: Seyde, Tim, et al.
Published: (2024)

AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning
by: Xi, Zhiheng, et al.
Published: (2025)

Normalization and effective learning rates in reinforcement learning
by: Lyle, Clare, et al.
Published: (2024)

Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making
by: Li, Manling, et al.
Published: (2024)

How do language models learn facts? Dynamics, curricula and hallucinations
by: Zucchet, Nicolas, et al.
Published: (2025)

How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition
by: Dong, Guanting, et al.
Published: (2023)

Position Paper: Rethinking Privacy in RL for Sequential Decision-making in the Age of LLMs
by: Fan, Flint Xiaofeng, et al.
Published: (2025)

PrunePEFT: Iterative Hybrid Pruning for Parameter-Efficient Fine-tuning of LLMs
by: Yu, Tongzhou, et al.
Published: (2025)

Maxwell's Demon at Work: Efficient Pruning by Leveraging Saturation of Neurons
by: Dufort-Labbé, Simon, et al.
Published: (2024)

Asynchronous Algorithmic Alignment with Cocycles
by: Dudzik, Andrew, et al.
Published: (2023)

Explaining RL Decisions with Trajectories
by: Deshmukh, Shripad Vilasrao, et al.
Published: (2023)

ZorBA: Zeroth-order Federated Fine-tuning of LLMs with Heterogeneous Block Activation
by: Meng, Chuiyang, et al.
Published: (2026)

Understanding Prompt Tuning and In-Context Learning via Meta-Learning
by: Genewein, Tim, et al.
Published: (2025)

PIANIST: Learning Partially Observable World Models with LLMs for Multi-Agent Decision Making
by: Light, Jonathan, et al.
Published: (2024)

Beyond Greedy Exits: Improved Early Exit Decisions for Risk Control and Reliability
by: Bajpai, Divya Jyoti, et al.
Published: (2025)

Torque-Aware Momentum
by: Malviya, Pranshu, et al.
Published: (2024)

Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset
by: Galashov, Alexandre, et al.
Published: (2024)

The Two-Stage Decision-Sampling Hypothesis: Understanding the Emergence of Self-Reflection in RL-Trained LLMs
by: Zhao, Zibo, et al.
Published: (2026)

Step-wise Adaptive Integration of Supervised Fine-tuning and Reinforcement Learning for Task-Specific LLMs
by: Chen, Jack, et al.
Published: (2025)

Quantized Evolution Strategies: High-precision Fine-tuning of Quantized LLMs at Low-precision Cost
by: Xu, Yinggan, et al.
Published: (2026)

Horizon Reduction Makes RL Scalable
by: Park, Seohong, et al.
Published: (2025)