Retrospective In-Context Learning for Temporal Credit Assignment with Large Language Models

• URL: http://arxiv.org/abs/2602.17497v1
• Date: Thu, 19 Feb 2026 16:13:28 GMT
• Title: Retrospective In-Context Learning for Temporal Credit Assignment with Large Language Models
• Authors: Wen-Tse Chen, Jiayu Chen, Fahim Tajwar, Hao Zhu, Xintong Duan, Ruslan Salakhutdinov, Jeff Schneider,
• Abstract summary: Temporal credit assignment mitigates this issue by transforming sparse feedback into dense supervision signals.<n>We propose to leverage pretrained knowledge from large language models (LLMs) to transform sparse rewards into dense training signals.<n>Online learning framework, RICOL, iteratively refines the policy based on the credit assignment results.
• Score: 41.59440766417004
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Learning from self-sampled data and sparse environmental feedback remains a fundamental challenge in training self-evolving agents. Temporal credit assignment mitigates this issue by transforming sparse feedback into dense supervision signals. However, previous approaches typically depend on learning task-specific value functions for credit assignment, which suffer from poor sample efficiency and limited generalization. In this work, we propose to leverage pretrained knowledge from large language models (LLMs) to transform sparse rewards into dense training signals (i.e., the advantage function) through retrospective in-context learning (RICL). We further propose an online learning framework, RICOL, which iteratively refines the policy based on the credit assignment results from RICL. We empirically demonstrate that RICL can accurately estimate the advantage function with limited samples and effectively identify critical states in the environment for temporal credit assignment. Extended evaluation on four BabyAI scenarios show that RICOL achieves comparable convergent performance with traditional online RL algorithms with significantly higher sample efficiency. Our findings highlight the potential of leveraging LLMs for temporal credit assignment, paving the way for more sample-efficient and generalizable RL paradigms.

Related papers

• MiniRec: Data-Efficient Reinforcement Learning for LLM-based Recommendation [50.417769112326546]
  MiniRec is a data selection framework tailored for RL-based large language models (LLMs) recommendation.<n>It evaluates sample learnability using key RL signals -- rewards -- pruning samples that are too easy (too high reward) or too difficult (consistently low reward)
  arXiv  Detail & Related papers  (2026-02-04T07:15:49Z)
• LANPO: Bootstrapping Language and Numerical Feedback for Reinforcement Learning in LLMs [73.27182315028021]
  LANPO is a framework that cleanly separates the roles of feedback: language guides exploration, while numerical rewards drive optimization.<n>Our work provides a robust method for integrating historical experiences into the LLM RL loop, creating more effective and data-efficient learning agents.
  arXiv  Detail & Related papers  (2025-10-18T15:51:19Z)
• Beyond Static LLM Policies: Imitation-Enhanced Reinforcement Learning for Recommendation [23.945049006150555]
  Large language models (LLMs) have become critical tools for enhancing user engagement by delivering personalized content across diverse digital platforms.<n>Direct deployment of LLMs as primary recommendation policies presents notable challenges, including persistent latency issues.<n>This paper proposes a novel offline reinforcement learning framework that leverages imitation learning from LLM-generated trajectories.
  arXiv  Detail & Related papers  (2025-10-15T07:28:29Z)
• CAPO: Towards Enhancing LLM Reasoning through Generative Credit Assignment [44.33395106709674]
  Reinforcement Learning with Verifiable Rewards (RLVR) has improved the reasoning abilities of Large Language Models (LLMs) by using rule-based binary feedback.<n>Current RLVR methods typically assign the same reward to every token.<n>This coarse-grained feedback hampers precise credit assignment, making it hard for models to identify which reasoning steps lead to success or failure.
  arXiv  Detail & Related papers  (2025-08-04T11:06:08Z)
• VinePPO: Refining Credit Assignment in RL Training of LLMs [66.80143024475635]
  We propose VinePPO, a straightforward approach that leverages the flexibility of language environments to compute unbiased Monte Carlo-based estimates.<n>Our method consistently outperforms PPO and other baselines across MATH and GSM8K datasets in less wall-clock time.
  arXiv  Detail & Related papers  (2024-10-02T15:49:30Z)
• How Can LLM Guide RL? A Value-Based Approach [68.55316627400683]
  Reinforcement learning (RL) has become the de facto standard practice for sequential decision-making problems by improving future acting policies with feedback. Recent developments in large language models (LLMs) have showcased impressive capabilities in language understanding and generation, yet they fall short in exploration and self-improvement capabilities. We develop an algorithm named LINVIT that incorporates LLM guidance as a regularization factor in value-based RL, leading to significant reductions in the amount of data needed for learning.
  arXiv  Detail & Related papers  (2024-02-25T20:07:13Z)
• Blending Imitation and Reinforcement Learning for Robust Policy Improvement [15.46204945792257]
  Imitation learning (IL) utilizes oracles to improve sample efficiency.<n>RPI draws on the strengths of IL, using oracle queries to facilitate exploration.<n>RPI is capable of learning from and improving upon a diverse set of black-box oracles.
  arXiv  Detail & Related papers  (2023-10-03T01:55:54Z)
• When does return-conditioned supervised learning work for offline reinforcement learning? [51.899892382786526]
  We study the capabilities and limitations of return-conditioned supervised learning. We find that RCSL returns the optimal policy under a set of assumptions stronger than those needed for the more traditional dynamic programming-based algorithms.
  arXiv  Detail & Related papers  (2022-06-02T15:05:42Z)

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.