REVIEW 7 cited by
HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Large Language Model (LLM)-based agents exhibit significant potential across various domains, operating as interactive systems that process environmental observations to generate executable actions for target tasks. The effectiveness of these agents is significantly influenced by their memory mechanism, which records historical experiences as sequences of action-observation pairs. We categorize memory into two types: cross-trial memory, accumulated across multiple attempts, and in-trial memory (working memory), accumulated within a single attempt. While considerable research has optimized performance through cross-trial memory, the enhancement of agent performance through improved working memory utilization remains underexplored. Instead, existing approaches often involve directly inputting entire historical action-observation pairs into LLMs, leading to redundancy in long-horizon tasks. Inspired by human problem-solving strategies, this paper introduces HiAgent, a framework that leverages subgoals as memory chunks to manage the working memory of LLM-based agents hierarchically. Specifically, HiAgent prompts LLMs to formulate subgoals before generating executable actions and enables LLMs to decide proactively to replace previous subgoals with summarized observations, retaining only the action-observation pairs relevant to the current subgoal. Experimental results across five long-horizon tasks demonstrate that HiAgent achieves a twofold increase in success rate and reduces the average number of steps required by 3.8. Additionally, our analysis shows that HiAgent consistently improves performance across various steps, highlighting its robustness and generalizability. Project Page: https://github.com/HiAgent2024/HiAgent .
Forward citations
Cited by 7 Pith papers
-
Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents
A tabular UCB controller trained on task success improves LLM-agent memory use over fixed heuristics, without extra LLM calls.
-
Consistent but Miscalibrated: Evaluating LLM Limitations for Risk Communication in Natural Language
Current LLMs produce consistent but miscalibrated natural-language descriptors of likelihood and uncertainty from probabilistic predictions and are not yet reliable zero-shot risk communicators.
-
ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL
Source-indexed selective turn memory plus provenance-guided credit assignment raises BrowseComp-Plus accuracy to 43.4% over GRPO and SUPO while cutting turn proliferation.
-
Beyond Hard and Soft: Hybrid Context Compression for Balancing Local and Global Information Retention
HyCo2 combines soft global compression with hard local token selection, reporting QA performance near uncompressed retrieval while cutting context tokens by about 88.8%.
-
Cost and Accuracy of Long-Term Memory in Distributed Multi-Agent Systems Based on Large Language Models
A two-framework testbed comparison claims mem0 is Pareto-optimal over Graphiti for distributed LLM agents because its lower cost is paired with accuracy that is not significantly different.
-
Symbolic Intermediaries as a Linguistic-Numerical Interface for LLM-Driven Geometric Reasoning
A dual-agent LLM loop with symbolic-regression feedback improved planar mechanism synthesis Chamfer distances in most tested settings, but the headline genetic-algorithm comparison and critique-analysis claims are not...
-
X-WebAgentBench: A Multilingual Interactive Web Benchmark for Evaluating Global Agentic System
X-WebAgentBench translates the WebShop e-commerce agent task into 14 languages and shows that current agents, including GPT-4o, perform substantially worse in multilingual settings than in English.
Discussion (0). Continue with ORCID to comment.