H$^2$R: Hierarchical Hindsight Reflection for Multi-Task LLM Agents

• URL: http://arxiv.org/abs/2509.12810v1
• Date: Tue, 16 Sep 2025 08:30:08 GMT
• Title: H$^2$R: Hierarchical Hindsight Reflection for Multi-Task LLM Agents
• Authors: Shicheng Ye, Chao Yu, Kaiqiang Ke, Chengdong Xu, Yinqi Wei,
• Abstract summary: Large language model (LLM)-based agents have shown strong potential in multi-task scenarios.<n>Existing approaches often treat prior experiences and knowledge as monolithic units, leading to inefficient and coarse-grained knowledge transfer.<n>We propose a novel hierarchical memory architecture that enables fine-grained knowledge transfer.
• Score: 3.9054156855794973
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language model (LLM)-based agents have shown strong potential in multi-task scenarios, owing to their ability to transfer knowledge across diverse tasks. However, existing approaches often treat prior experiences and knowledge as monolithic units, leading to inefficient and coarse-grained knowledge transfer. In this work, we propose a novel hierarchical memory architecture that enables fine-grained knowledge transfer by decoupling high-level planning memory from low-level execution memory. To construct and refine these hierarchical memories, we introduce Hierarchical Hindsight Reflection (H$^2$R), a mechanism that distills reusable and hierarchical knowledge from past agent-environment interactions. At test time, H$^2$R performs retrievals of high-level and low-level memories separately, allowing LLM-based agents to efficiently access and utilize task-relevant knowledge for new tasks.Experimental results across two benchmarks demonstrate that H$^2$R can improve generalization and decision-making performance, outperforming prior baselines such as Expel.

Related papers

• Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory [89.65731902036669]
  Evo-Memory is a streaming benchmark and framework for evaluating self-evolving memory in large language model (LLM) agents.<n>We evaluate over ten representative memory modules and evaluate them across 10 diverse multi-turn goal-oriented and single-turn reasoning and QA datasets.
  arXiv  Detail & Related papers  (2025-11-25T21:08:07Z)
• A Benchmark for Procedural Memory Retrieval in Language Agents [0.023227405857540805]
  Current AI agents excel in familiar settings, but fail sharply when faced with novel tasks with unseenProc.<n>We present the first benchmark that isolates procedural memory retrieval from task execution.<n>Our results expose a clear generalization cliff: embedding-based methods perform strongly on familiar contexts, yet degrade considerably on novel ones.
  arXiv  Detail & Related papers  (2025-11-21T08:08:53Z)
• RECALL: REpresentation-aligned Catastrophic-forgetting ALLeviation via Hierarchical Model Merging [33.22889542330089]
  Internal representations in large language models (LLMs) serve as reliable proxies of learned knowledge.<n>We propose RECALL, a representation-aware model merging framework for continual learning without access to historical data.
  arXiv  Detail & Related papers  (2025-10-23T12:17:37Z)
• ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory [57.517214479414726]
  ReasoningBank is a memory framework that distills generalizable reasoning strategies from an agent's self-judged successful and failed experiences.<n>At test time, an agent retrieves relevant memories from ReasoningBank to inform its interaction and then integrates new learnings back, enabling it to become more capable over time.<n>We introduce memory-aware test-time scaling (MaTTS), which accelerates and diversifies this learning process by scaling up the agent's interaction experience.
  arXiv  Detail & Related papers  (2025-09-29T17:51:03Z)
• Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning [59.16831804985279]
  Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of NLP tasks, but they remain fundamentally stateless.<n>Recent efforts to address this limitation often augment LLMs with an external memory bank, yet most existing pipelines are static and learned.<n>We present Memory-R1, a reinforcement learning framework that equips LLMs with the ability to actively manage and utilize external memory.
  arXiv  Detail & Related papers  (2025-08-27T12:26:55Z)
• Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents [19.04968632268433]
  We propose a hierarchical memory architecture for Large Language Model Agents (LLM Agents)<n>Each memory vector is embedded with a positional index encoding pointing to its semantically related sub-memories in the next layer.<n>During the reasoning phase, an index-based routing mechanism enables efficient, layer-by-layer retrieval without performing exhaustive similarity computations.
  arXiv  Detail & Related papers  (2025-07-23T12:45:44Z)
• Re-ranking Reasoning Context with Tree Search Makes Large Vision-Language Models Stronger [51.01841635655944]
  Recent advancements in Large Vision Language Models (LVLMs) have significantly improved performance in Visual Question Answering (VQA) tasks.<n>Existing methods still face challenges, such as the scarcity of knowledge with reasoning examples and erratic responses from retrieved knowledge.<n>We propose a multimodal RAG framework, termed RCTS, which enhances LVLMs by constructing a Reasoning Context-enriched knowledge base and a Tree Search re-ranking method.
  arXiv  Detail & Related papers  (2025-06-09T14:00:57Z)
• Learning to Route Queries Across Knowledge Bases for Step-wise Retrieval-Augmented Reasoning [60.84901522792042]
  Multimodal Retrieval-Augmented Generation (MRAG) has shown promise in mitigating hallucinations in Multimodal Large Language Models (MLLMs)<n>We propose R1, a novel MRAG framework that learns to decide when and where to retrieve knowledge based on the evolving reasoning state.<n>R1- can adaptively and effectively leverage diverse KBs, reducing unnecessary retrievals and improving both efficiency and accuracy.
  arXiv  Detail & Related papers  (2025-05-28T08:17:57Z)
• Efficiently Enhancing General Agents With Hierarchical-categorical Memory [0.5919433278490629]
  We introduce EHC, a general agent capable of learning without parameter updates.<n>EHC consists of a Hierarchical Memory Retrieval (HMR) module and a Task-Category Oriented Experience Learning (TOEL) module.
  arXiv  Detail & Related papers  (2025-05-28T06:12:51Z)
• From RAG to Memory: Non-Parametric Continual Learning for Large Language Models [6.380729797938521]
  retrieval-augmented generation (RAG) has become the dominant way to introduce new information.<n>Recent RAG approaches augment vector embeddings with various structures like knowledge graphs to address some gaps, namely sense-making and associativity.<n>We propose HippoRAG 2, a framework that outperforms standard RAG comprehensively on factual, sense-making, and associative memory tasks.
  arXiv  Detail & Related papers  (2025-02-20T18:26:02Z)
• mR$^2$AG: Multimodal Retrieval-Reflection-Augmented Generation for Knowledge-Based VQA [78.45521005703958]
  multimodal Retrieval-Augmented Generation (mRAG) is naturally introduced to provide MLLMs with comprehensive and up-to-date knowledge. We propose a novel framework called textbfRetrieval-textbfReftextbfAugmented textbfGeneration (mR$2$AG) which achieves adaptive retrieval and useful information localization. mR$2$AG significantly outperforms state-of-the-art MLLMs on INFOSEEK and Encyclopedic-VQA
  arXiv  Detail & Related papers  (2024-11-22T16:15:50Z)
• MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentation [60.04380907045708]
  Retrieval-Augmented Generation (RAG) is considered a promising strategy to address this problem.<n>We propose MemoRAG, a novel RAG framework empowered by global memory-augmented retrieval.<n>MemoRAG achieves superior performances across a variety of long-context evaluation tasks.
  arXiv  Detail & Related papers  (2024-09-09T13:20:31Z)

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.