14 citations · 31 across the 22 of their papers we have counts for
24 papers
Skill Use or Skill Theater? Evaluating the Reasoning Backroom in Skill-Augmented Language Agents
Jinwei Hu, Yi Qi, Xinmiao Huang +3
Reusable skills are becoming a standard interface for extending language agents with task procedures. Yet evaluators usually infer skill use from visible reasoning or the agent's o…
SkillFuzz: Fuzzing Skill Composition for Implicit Intents Discovery in Open Skill Marketplaces
Jinwei Hu, Yi Dong, Youcheng Sun +1
Large Language Model (LLM)-based agents increasingly automate software engineering tasks through reusable skills, natural-language instruction documents that guide planning and exe…
Responsible Agentic AI Requires Explicit Provenance
Jinwei Hu, Xinmiao Huang, Qisong He +3
Agentic AI is rapidly proliferating across diverse real-world domains such as software engineering, yet public trust has not kept pace. The central reason is that responsibility, d…
Safety-Constrained Reinforcement Learning with Post-Training Reachability Verification for Robot Navigation
Qisong He, Xinmiao Huang, Jinwei Hu +4
Safe navigation for mobile robots demands policies that remain reliable under the high-consequence perception uncertainty of cluttered environments. Yet most existing safe reinforc…
PrefixGuard: Online Failure Warning and Trace-Grounded Diagnosis for LLM Agents
Xinmiao Huang, Jinwei Hu, Rajarshi Roy +4
Large language model (LLM) agents execute multi-step tasks, but terminal verdicts arrive too late for intervention. Agent traces mix messages, tool calls, and feedback, making it d…
Where Do Prompt Perturbations Break Generation? A Segment-Level View of Robustness in LoRA-Tuned Language Models
Zhuoyun Li, Boxuan Wang, Jinwei Hu +6
Large language models are sensitive to minor prompt perturbations, yet existing robustness methods usually enforce consistency at the whole-sequence level. This holistic view can h…
FragileFlow: Spectral Control of Correct-but-Fragile Predictions for Foundation Model Robustness
Zhuoyun Li, Boxuan Wang, Jinwei Hu +2
Robust adaptation of LLMs and VLMs is often evaluated by average accuracy or average consistency under perturbations. However, these averages can hide a structured failure mode: a…
Spatiotemporal System Forecasting with Irregular Time Steps via Masked Autoencoder
Kewei Zhu, Yanze Xin, Jinwei Hu +3
Predicting high-dimensional dynamical systems with irregular time steps presents significant challenges for current data-driven algorithms. These irregularities arise from missing…
DDSA: Dual-Domain Strategic Attack for Spatial-Temporal Efficiency in Adversarial Robustness Testing
Jinwei Hu, Shiyuan Meng, Yi Dong +1
Image transmission and processing systems in resource-critical applications face significant challenges from adversarial perturbations that compromise mission-specific object class…
Lying with Truths: Open-Channel Multi-Agent Collusion for Belief Manipulation via Generative Montage
Jinwei Hu, Xinmiao Huang, Youcheng Sun +2
As large language models (LLMs) transition to autonomous agents synthesizing real-time information, their reasoning capabilities introduce an unexpected attack surface. This paper…
Stop Reducing Responsibility in LLM-Powered Multi-Agent Systems to Local Alignment
Jinwei Hu, Yi Dong, Shuang Ao +6
LLM-powered Multi-Agent Systems (LLM-MAS) unlock new potentials in distributed reasoning, collaboration, and task generalization but also introduce additional risks due to unguaran…
Tapas Are Free! Training-Free Adaptation of Programmatic Agents via LLM-Guided Program Synthesis in Dynamic Environments
Jinwei Hu, Yi Dong, Youcheng Sun +1
Autonomous agents in safety-critical applications must continuously adapt to dynamic conditions without compromising performance and reliability. This work introduces TAPA (Trainin…
Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing
Jinwei Hu, Yi Dong, Zhengtao Ding +1
This paper presents a defense framework for enhancing the safety of large language model (LLM) empowered multi-agent systems (MAS) in safety-critical domains such as aerospace. We…
Hierarchical Testing with Rabbit Optimization for Industrial Cyber-Physical Systems
Jinwei Hu, Zezhi Tang, Xin Jin +3
This paper presents HERO (Hierarchical Testing with Rabbit Optimization), a novel black-box adversarial testing framework for evaluating the robustness of deep learning-based Progn…
Safe Pruning LoRA: Robust Distance-Guided Pruning for Safety Alignment in Adaptation of LLMs
Shuang Ao, Yi Dong, Jinwei Hu +1
Fine-tuning Large Language Models (LLMs) with Low-Rank Adaptation (LoRA) enhances adaptability while reducing computational costs. However, fine-tuning can compromise safety alignm…
TAIJI: Textual Anchoring for Immunizing Jailbreak Images in Vision Language Models
Xiangyu Yin, Yi Qi, Jinwei Hu +5
Vision Language Models (VLMs) have demonstrated impressive inference capabilities, but remain vulnerable to jailbreak attacks that can induce harmful or unethical responses. Existi…
CeTAD: Towards Certified Toxicity-Aware Distance in Vision Language Models
Xiangyu Yin, Jiaxu Liu, Zhen Chen +4
Recent advances in large vision-language models (VLMs) have demonstrated remarkable success across a wide range of visual understanding tasks. However, the robustness of these mode…
Position: Towards a Responsible LLM-empowered Multi-Agent Systems
Jinwei Hu, Yi Dong, Shuang Ao +6
The rise of Agent AI and Large Language Model-powered Multi-Agent Systems (LLM-MAS) has underscored the need for responsible and dependable system operation. Tools like LangChain a…
FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model
Jinwei Hu, Zhenglin Huang, Xiangyu Yin +4
Large language models have been widely applied, but can inadvertently encode sensitive or harmful information, raising significant safety concerns. Machine unlearning has emerged t…
SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model
Zhenglin Huang, Jinwei Hu, Xiangtai Li +6
The rapid advancement of generative models in creating highly realistic images poses substantial risks for misinformation dissemination. For instance, a synthetic image, when share…
Trust-Oriented Adaptive Guardrails for Large Language Models
Jinwei Hu, Yi Dong, Xiaowei Huang
Guardrail, an emerging mechanism designed to ensure that large language models (LLMs) align with human values by moderating harmful or toxic responses, requires a sociotechnical ap…
Safeguarding Large Language Models: A Survey
Yi Dong, Ronghui Mu, Yanghao Zhang +9
In the burgeoning field of Large Language Models (LLMs), developing a robust safety mechanism, colloquially known as "safeguards" or "guardrails", has become imperative to ensure t…
Explainable Global Wildfire Prediction Models using Graph Neural Networks
Dayou Chen, Sibo Cheng, Jinwei Hu +2
Wildfire prediction has become increasingly crucial due to the escalating impacts of climate change. Traditional CNN-based wildfire prediction models struggle with handling missing…
Building Guardrails for Large Language Models
Yi Dong, Ronghui Mu, Gaojie Jin +6
As Large Language Models (LLMs) become more integrated into our daily lives, it is crucial to identify and mitigate their risks, especially when the risks can have profound impacts…