16 papers
WASP-Mem: Let Agent Organize Procedural Memory. Human Spend Effort on Diagnosis
Peizhi Niu, Ge Shi, Vaibhav Gorde +4
LLM agents struggle to automate complex enterprise procedures because static, pre-built harness cannot capture proprietary policies, undocumented behaviors, or state-dependent reco…
Making LLMs Truly Forget: Deep Unlearning by Searching, Selecting, and Severing Knowledge Paths
Jialu Wang, Peizhi Niu, Haoteng Yin +3
While an unlearned language model may no longer recall a fact directly, the fact often remains recoverable through multi-hop reasoning over related knowledge. Most existing unlearn…
What Must a World Model Distinguish for Planning?
Rongzhe Wei, Hans Hao-Hsun Hsu, Peizhi Niu +2
World models simulate the consequences of action candidates, but good planning need not preserve every physical distinction required for accurate prediction. We formalize this gap…
Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens
Peizhi Niu, Wenjie Qu, Shangding Gu +14
Claw-like AI agents (e.g., OpenClaw) are always-on processes with persistent access to credentials, files, tools, and external services. They take on system-level responsibilities…
MCP-Persona: Benchmarking LLM Agents on Real-World Personal Applications via Environment Simulation
Wenhao Wang, Peizhi Niu, Gongyi Zou +9
The Model Context Protocol (MCP) has emerged as a transformative standard for connecting large language models (LLMs) with external data sources and tools, and has been rapidly ado…
Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents
Ahmad Al-Tawaha, Shangding Gu, Peizhi Niu +2
Safety evaluations of memory-equipped LLM agents typically measure within-task safety: whether an agent completes a single scenario safely, often under adversarial conditions such…
One Turn Too Late: Learning When to Intervene Against Multi-Turn Malicious Intent
Xinjie Shen, Rongzhe Wei, Peizhi Niu +6
Hidden malicious intent in multi-turn dialogue poses a growing threat to deployed large language models (LLMs). Rather than exposing a harmful objective in a single prompt, attacke…
AgentSPEX: An Agent SPecification and EXecution Language
Pengcheng Wang, Jerry Huang, Jiarui Yao +7
Language-model agent systems commonly rely on reactive prompting, in which a single instruction guides the model through an open-ended sequence of reasoning and tool-use steps, lea…
The Trojan Knowledge: Bypassing Commercial LLM Guardrails via Harmless Prompt Weaving and Adaptive Tree Search
Rongzhe Wei, Peizhi Niu, Xinjie Shen +7
Large language models (LLMs) remain vulnerable to jailbreak attacks that bypass safety guardrails to elicit harmful outputs. Existing approaches overwhelmingly operate within the p…
MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools
Wenhao Wang, Peizhi Niu, Zhao Xu +8
Large Language Models (LLMs) increasingly rely on external tools to perform complex, realistic tasks, yet their ability to utilize the rapidly expanding Model Contextual Protocol (…
3DGS-IEval-15K: A Large-scale Image Quality Evaluation Database for 3D Gaussian-Splatting
Yuke Xing, Jiarui Wang, Peizhi Niu +3
3D Gaussian Splatting (3DGS) has emerged as a promising approach for novel view synthesis, offering real-time rendering with high visual fidelity. However, its substantial storage…
GUARD: Guided Unlearning and Retention via Data Attribution for Large Language Models
Peizhi Niu, Evelyn Ma, Huiting Zhou +5
Unlearning large language models (LLMs) is important due to copyright and privacy concerns. A key challenge is unintended forgetting, where removing data inadvertently impairs mode…
Do LLMs Really Forget? Evaluating Unlearning with Knowledge Correlation and Confidence Awareness
Rongzhe Wei, Peizhi Niu, Hans Hao-Hsun Hsu +9
Machine unlearning techniques aim to mitigate unintended memorization in large language models (LLMs). However, existing approaches predominantly focus on the explicit removal of i…
DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platform
Peizhi Niu, Yu-Hsiang Wang, Vishal Rana +3
We introduce a new graph diffusion model for small molecule generation, DMol, which outperforms the state-of-the-art DiGress model in terms of validity by roughly 1.5% across all b…
ModelVerification.jl: a Comprehensive Toolbox for Formally Verifying Deep Neural Networks
Tianhao Wei, Hanjiang Hu, Luca Marzari +4
Deep Neural Networks (DNN) are crucial in approximating nonlinear functions across diverse applications, ranging from image classification to control. Verifying specific input-outp…
Graph Transductive Defense: a Two-Stage Defense for Graph Membership Inference Attacks
Peizhi Niu, Chao Pan, Siheng Chen +1
Graph neural networks (GNNs) have become instrumental in diverse real-world applications, offering powerful graph learning capabilities for tasks such as social networks and medica…