14 papers
CoEvolve: Counterfactual-Guided Co-Evolution of Memory and Collaboration in Multi-Agent Systems
Xufei Shi, Derong Xu, Wei Chen +3
Large language model (LLM)-based multi-agent systems have demonstrated remarkable capabilities across diverse tasks. Yet their memory and collaboration architectures often remain s…
ATLAS: Dual-Horizon Diagnostic Evaluation for Industrial Tool-Use Agents
Wei Chen, Peilun Zhou, Zhaoyu Hu +8
Large language model (LLM) agents are increasingly deployed in user-facing services that require iterative tool use under dynamic business conditions. Reliable evaluation is essent…
SAFE-G: Structure-aware Faithful Evidence-guided Generation for Knowledge-based Visual Question Answering
Long Shu, Shuochen Liu, Wei Chen +4
Knowledge-based Visual Question Answering (KB-VQA) aims to answer queries that necessitate reasoning over external knowledge sources beyond the visual content. Typically, current m…
Unlocking Parallelism in Autoregressive Language Models via Speculative Decoding with Progressive Tree Drafting
Zipeng Gao, Zhi Zheng, Qingrong Xia +5
Speculative decoding has significantly accelerated Large Language Model (LLM) inference by alleviating memory-bound bottlenecks. However, traditional speculative decoding typically…
From Blueprint to Reality: Modeling and Applying Putnam's Social Capital Theory with LLM-based Multi-agent Simulations
Shiyi Ling, Zhi Zheng, Hui Zheng +3
Putnam's Social Capital Theory is a foundational framework for collective action and community prosperity. However, traditional empirical methods face practical limits on control a…
AtomMem: Building Simple and Effective Memory System for LLM Agents via Atomic Facts
Yanyu Yao, Shangze Li, Zhi Zheng +4
Large language models (LLMs) demonstrate strong reasoning and generation abilities, but their fixed context windows limit long-term information accumulation and reuse across multi-…
SmellBench: Towards Fine-Grained Evaluation of Code Agents on Refactoring Tasks
Fake Lin, Binbin Hu, Xi Zhu +6
Code Agents have achieved remarkable advances in recent years, exhibiting strong capabilities across a wide range of software engineering tasks. However, their misuse often produce…
Entropy-KL Divergence-based Token Masking: A Novel Approach for Selective Fine-tuning of Large Language Models
Qi Liu, Mingdi Sun, Yongyi He +5
Supervised fine-tuning (SFT) followed by reinforcement learning (RL) has become a standard post-training paradigm for large language models. This paradigm provides a cold-start for…
Defending LLM-based Multi-Agent Systems Against Cooperative Attacks with Sentence-Level Rectification
Yaoyang Luo, Zhi Zheng, Ziwei Zhao +5
Recent years have witnessed the rapid development of Large Language Model-based Multi-Agent Systems (MAS), which excel at collaborative decision-making and complex problem-solving.…
EviDep: Uncertainty-Aware Multimodal Depression Estimation via Disentangled Evidential Learning
Fangyuan Liu, Sirui Zhao, Yangsong Zhang +7
Audio--visual recordings provide complementary cues for estimating depression severity, but their informativeness varies across time and modalities. Point predictions alone do not…
More Edits, More Stable: Understanding the Lifelong Normalization in Sequential Model Editing
Xin Ma, Wei Chen, Qi Liu +4
Lifelong Model Editing aims to continuously update evolving facts in Large Language Models while preserving unrelated knowledge and general capabilities, yet it remains plagued by…
STRIDE: Strategic Iterative Decision-Making for Retrieval-Augmented Multi-Hop Question Answering
Wei Chen, Lili Zhao, Zhi Zheng +2
Multi-hop question answering (MHQA) enables accurate answers to complex queries by retrieving and reasoning over evidence dispersed across multiple documents. Existing MHQA approac…
VIGIL: Defending LLM Agents Against Tool Stream Injection via Verify-Before-Commit
Junda Lin, Zhaomeng Zhou, Zhi Zheng +4
LLM agents operating in open environments face escalating risks from indirect prompt injection, particularly within the tool stream where manipulated metadata and runtime feedback…
DynaDebate: Breaking Homogeneity in Multi-Agent Debate with Dynamic Path Generation
Zhenghao Li, Zhi Zheng, Wei Chen +4
Recent years have witnessed the rapid development of Large Language Model-based Multi-Agent Systems (MAS), which excel at collaborative decision-making and complex problem-solving.…