18 citations · 20 across the 33 of their papers we have counts for
34 papers
Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
Wenbo Pan, Shujie Liu, Chin-Yew Lin +5
AI agents rely on a harness of skills, tools, and workflows to solve complex problems. Continually improving this harness is essential for adapting to new tasks. However, existing…
From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models
Juncheng Wu, Hardy Chen, Haoqin Tu +6
Recent advances in vision-language models (VLMs) emphasize long chain-of-thought reasoning; yet, we find that their performance on visual tasks is primarily limited by a lack of vi…
Firefly: Illuminating Large-Scale Verified Tool-Call Data Generation from Real APIs
Yuxuan Lu, Ziyi Wang, Yingzhou Lu +12
Training tool-calling agents requires large-scale trajectory data with verifiable labels, yet existing approaches either synthesize environments that diverge from real API behavior…
Position: Agentic Evolution is the Path to Evolving LLMs
Minhua Lin, Hanqing Lu, Zhan Shi +11
As Large Language Models (LLMs) move from curated training sets into open-ended real-world environments, a fundamental limitation emerges: static training cannot keep pace with con…
MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution
Minhua Lin, Zhiwei Zhang, Hanqing Lu +5
Memory-augmented LLM agents maintain external memory banks to support long-horizon interaction, yet most existing systems treat construction, retrieval, and utilization as isolated…
How Far Are LLMs from Professional Poker Players? Revisiting Game-Theoretic Reasoning with Agentic Tool Use
Minhua Lin, Enyan Dai, Hui Liu +11
As Large Language Models (LLMs) are increasingly applied in high-stakes domains, their ability to reason strategically under uncertainty becomes critical. Poker provides a rigorous…
Trajectory2Task: Training Robust Tool-Calling Agents with Synthesized Yet Verifiable Data for Complex User Intents
Ziyi Wang, Yuxuan Lu, Yimeng Zhang +12
Tool-calling agents are increasingly deployed in real-world customer-facing workflows. Yet most studies on tool-calling agents focus on idealized settings with general, fixed, and…
Unlocking the Power of Multi-Agent LLM for Reasoning: From Lazy Agents to Deliberation
Zhiwei Zhang, Xiaomin Li, Yudi Lin +8
Large Language Models (LLMs) trained with reinforcement learning and verifiable rewards have achieved strong results on complex reasoning tasks. Recent work extends this paradigm t…
Generalizing Test-time Compute-optimal Scaling as an Optimizable Graph
Fali Wang, Jihai Chen, Shuhua Yang +7
Test-Time Scaling (TTS) improves large language models (LLMs) by allocating additional computation during inference, typically through parallel, sequential, or hybrid scaling. Howe…
A Comprehensive Survey on Reinforcement Learning-based Agentic Search: Foundations, Roles, Optimizations, Evaluations, and Applications
Minhua Lin, Zongyu Wu, Zhichao Xu +6
The advent of large language models (LLMs) has transformed information access and reasoning through open-ended natural language interaction. However, LLMs remain limited by static…
Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs
Zhining Liu, Ziyi Chen, Hui Liu +9
Vision-Language Models (VLMs) achieve strong results on multimodal tasks such as visual question answering, yet they can still fail even when the correct visual evidence is present…
TRAJECT-Bench:A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use
Pengfei He, Zhenwei Dai, Bing He +10
Large language model (LLM)-based agents increasingly rely on tool use to complete real-world tasks. While existing works evaluate the LLMs' tool use capability, they largely focus…
Adaptive Test-Time Reasoning via Reward-Guided Dual-Phase Search
Yingqian Cui, Zhenwei Dai, Pengfei He +8
Large Language Models (LLMs) have achieved significant advances in reasoning tasks. A key approach is tree-based search with verifiers, which expand candidate reasoning paths and u…
Bradley-Terry and Multi-Objective Reward Modeling Are Complementary
Zhiwei Zhang, Hui Liu, Xiaomin Li +10
Reward models trained on human preference data have demonstrated strong effectiveness in aligning Large Language Models (LLMs) with human intent under the framework of Reinforcemen…
Efficient Long CoT Reasoning in Small Language Models
Zhaoyang Wang, Jinqi Jiang, Tian Qiu +3
Recent large reasoning models such as DeepSeek-R1 exhibit strong complex problems solving abilities by generating long chain-of-thought (CoT) reasoning steps. It is challenging to…
SUA: Stealthy Multimodal Large Language Model Unlearning Attack
Xianren Zhang, Hui Liu, Delvin Ce Zhang +4
Multimodal Large Language Models (MLLMs) trained on massive data may memorize sensitive personal information and photos, posing serious privacy risks. To mitigate this, MLLM unlear…
To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems
Pengfei He, Zhenwei Dai, Xianfeng Tang +9
Large Language Model-based Multi-Agent Systems (LLM-MAS) have demonstrated strong capabilities in solving complex tasks but remain vulnerable when agents receive unreliable message…
Comprehensive Vulnerability Analysis is Necessary for Trustworthy LLM-MAS
Pengfei He, Yue Xing, Juanhui Li +7
TThis paper argues that \textbf{a comprehensive vulnerability analysis is essential for building trustworthy Large Language Model-based Multi-Agent Systems (LLM-MAS)}. These system…
Keeping an Eye on LLM Unlearning: The Hidden Risk and Remedy
Jie Ren, Zhenwei Dai, Xianfeng Tang +9
Although Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of tasks, growing concerns have emerged over the misuse of sensitive, copyrighte…
Beyond Text: Unveiling Privacy Vulnerabilities in Multi-modal Retrieval-Augmented Generation
Jiankun Zhang, Shenglai Zeng, Jie Ren +4
Multimodal Retrieval-Augmented Generation (MRAG) systems enhance LMMs by integrating external multimodal databases, but introduce unexplored privacy vulnerabilities. While text-bas…
SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models
Hardy Chen, Haoqin Tu, Fali Wang +5
This work revisits the dominant supervised fine-tuning (SFT) then reinforcement learning (RL) paradigm for training Large Vision-Language Models (LVLMs), and reveals a key finding:…
Harnessing the Unseen: The Hidden Influence of Intrinsic Knowledge in Long-Context Language Models
Yu Fu, Haz Sameen Shahgir, Hui Liu +3
Recent advances in long-context language models (LCLMs), designed to handle extremely long contexts, primarily focus on utilizing external contextual information, often leaving the…
ViLBench: A Suite for Vision-Language Process Reward Modeling
Haoqin Tu, Weitao Feng, Hardy Chen +3
Process-supervised reward models serve as a fine-grained function that provides detailed step-wise feedback to model responses, facilitating effective selection of reasoning trajec…
A General Framework to Enhance Fine-tuning-based LLM Unlearning
Jie Ren, Zhenwei Dai, Xianfeng Tang +7
Unlearning has been proposed to remove copyrighted and privacy-sensitive data from Large Language Models (LLMs). Existing approaches primarily rely on fine-tuning-based methods, wh…