Safe In-Context Reinforcement Learning

• URL: http://arxiv.org/abs/2509.25582v1
• Date: Mon, 29 Sep 2025 23:07:32 GMT
• Title: Safe In-Context Reinforcement Learning
• Authors: Amir Moeini, Minjae Kwon, Alper Kamil Bozkurt, Yuichi Motai, Rohan Chandra, Lu Feng, Shangtong Zhang,
• Abstract summary: In-context reinforcement learning (ICRL) is an emerging RL paradigm where the agent, after some pretraining procedure, is able to adapt to out-of-distribution test tasks without any parameter updates.<n>We propose the first method that promotes the safety of ICRL's adaptation process in the framework of constrained Markov Decision Processes.
• Score: 19.11554873825587
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In-context reinforcement learning (ICRL) is an emerging RL paradigm where the agent, after some pretraining procedure, is able to adapt to out-of-distribution test tasks without any parameter updates. The agent achieves this by continually expanding the input (i.e., the context) to its policy neural networks. For example, the input could be all the history experience that the agent has access to until the current time step. The agent's performance improves as the input grows, without any parameter updates. In this work, we propose the first method that promotes the safety of ICRL's adaptation process in the framework of constrained Markov Decision Processes. In other words, during the parameter-update-free adaptation process, the agent not only maximizes the reward but also minimizes an additional cost function. We also demonstrate that our agent actively reacts to the threshold (i.e., budget) of the cost tolerance. With a higher cost budget, the agent behaves more aggressively, and with a lower cost budget, the agent behaves more conservatively.

Related papers

• Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates [53.3717573880076]
  We introduce Just-In-Time Reinforcement Learning (JitRL), a training-free framework that enables test-time policy optimization without any gradient updates.<n>JitRL maintains a dynamic, non-parametric memory of experiences and retrieves relevant trajectories to estimate action advantages on-the-fly.<n>Experiments on WebArena and Jericho demonstrate that JitRL establishes a new state-of-the-art among training-free methods.
  arXiv  Detail & Related papers  (2026-01-26T14:16:51Z)
• SCOPE: Prompt Evolution for Enhancing Agent Effectiveness [53.75986399936395]
  Large Language Model (LLM) agents are increasingly deployed in environments that generate massive, dynamic contexts.<n>While agents have access to this context, their static prompts lack the mechanisms to manage it effectively.<n>We introduce textbfSCOPE (Self-evolving Context Optimization via Prompt Evolution)<n>We propose a Dual-Stream mechanism that balances tactical specificity (resolving immediate errors) with strategic generality (evolving long-term principles)
  arXiv  Detail & Related papers  (2025-12-17T12:25:05Z)
• Budget-Aware Tool-Use Enables Effective Agent Scaling [82.6942342482552]
  Scaling test-time computation improves performance across different tasks on large language models (LLMs)<n>We study how to scale such agents effectively under explicit tool-call budgets, focusing on web search agents.<n>We introduce the Budget Tracker, a lightweight plug-in that provides the agent with continuous budget awareness.
  arXiv  Detail & Related papers  (2025-11-21T07:18:55Z)
• AgentPRM: Process Reward Models for LLM Agents via Step-Wise Promise and Progress [71.02263260394261]
  Large language models (LLMs) still encounter challenges in multi-turn decision-making tasks.<n>We build process reward models (PRMs) to evaluate each decision and guide the agent's decision-making process.<n>AgentPRM captures both the interdependence between sequential decisions and their contribution to the final goal.
  arXiv  Detail & Related papers  (2025-11-11T14:57:54Z)
• Towards Provable Emergence of In-Context Reinforcement Learning [20.54963042400693]
  Some reinforcement learning (RL) agents can solve a wide range of new out-of-distribution tasks without parameter updates after pretraining on some task distribution.<n>When evaluated in a new task, instead of making parameter updates, the pretrained agent conditions its policy on additional input called the context.<n>This phenomenon is typically called in-context RL (ICRL)
  arXiv  Detail & Related papers  (2025-09-22T20:23:09Z)
• $Agent^2$: An Agent-Generates-Agent Framework for Reinforcement Learning Automation [5.325886106098561]
  Reinforcement learning (RL) agent development traditionally requires substantial expertise and iterative effort.<n>This paper introduces Agent$2$, an LLM-driven agent-generates-agent framework for fully automated RL agent design.<n>Agent$2$ translates natural language task descriptions and environment code into executable RL solutions without human intervention.
  arXiv  Detail & Related papers  (2025-09-16T02:14:39Z)
• Memento: Fine-tuning LLM Agents without Fine-tuning LLMs [36.3424780932712]
  We introduce a novel learning paradigm for Adaptive Large Language Model (LLM) agents.<n>Our method enables low-cost continual adaptation via memory-based online reinforcement learning.<n>We instantiate our agent model in the deep research setting, namely emphMemento, which attains top-1 on GAIA validation.
  arXiv  Detail & Related papers  (2025-08-22T07:25:30Z)
• The Real Barrier to LLM Agent Usability is Agentic ROI [110.31127571114635]
  Large Language Model (LLM) agents represent a promising shift in human-AI interaction.<n>We highlight a critical usability gap in high-demand, mass-market applications.
  arXiv  Detail & Related papers  (2025-05-23T11:40:58Z)
• Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments [54.67512489842682]
  Large language models (LLMs) have demonstrated strong planning and decision-making capabilities in complex embodied environments.<n>We take a first step toward exploring the early-exit behavior for LLM-based agents.
  arXiv  Detail & Related papers  (2025-05-23T08:23:36Z)
• Decentralized scheduling through an adaptive, trading-based multi-agent system [1.7403133838762448]
  In multi-agent reinforcement learning systems, the actions of one agent can have a negative impact on the rewards of other agents. This work applies a trading approach to a simulated scheduling environment, where the agents are responsible for the assignment of incoming jobs to compute cores. The agents can trade the usage right of computational cores to process high-priority, high-reward jobs faster than low-priority, low-reward jobs.
  arXiv  Detail & Related papers  (2022-07-05T13:50:18Z)
• Differential Assessment of Black-Box AI Agents [29.98710357871698]
  We propose a novel approach to differentially assess black-box AI agents that have drifted from their previously known models. We leverage sparse observations of the drifted agent's current behavior and knowledge of its initial model to generate an active querying policy. Empirical evaluation shows that our approach is much more efficient than re-learning the agent model from scratch.
  arXiv  Detail & Related papers  (2022-03-24T17:48:58Z)
• Retrieval-Augmented Reinforcement Learning [63.32076191982944]
  We train a network to map a dataset of past experiences to optimal behavior. The retrieval process is trained to retrieve information from the dataset that may be useful in the current context. We show that retrieval-augmented R2D2 learns significantly faster than the baseline R2D2 agent and achieves higher scores.
  arXiv  Detail & Related papers  (2022-02-17T02:44:05Z)
• Modeling Bounded Rationality in Multi-Agent Simulations Using Rationally Inattentive Reinforcement Learning [85.86440477005523]
  We study more human-like RL agents which incorporate an established model of human-irrationality, the Rational Inattention (RI) model. RIRL models the cost of cognitive information processing using mutual information. We show that using RIRL yields a rich spectrum of new equilibrium behaviors that differ from those found under rational assumptions.
  arXiv  Detail & Related papers  (2022-01-18T20:54:00Z)
• Plan Better Amid Conservatism: Offline Multi-Agent Reinforcement Learning with Actor Rectification [74.10976684469435]
  offline reinforcement learning (RL) algorithms can be transferred to multi-agent settings directly. We propose a simple yet effective method, Offline Multi-Agent RL with Actor Rectification (OMAR), to tackle this critical challenge. OMAR significantly outperforms strong baselines with state-of-the-art performance in multi-agent continuous control benchmarks.
  arXiv  Detail & Related papers  (2021-11-22T13:27:42Z)

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

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.