1k citations · 1.1k across the 35 of their papers we have counts for
40 papers
AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned Prefixes
Zhuoxuan Zhang, Jinhao Duan, Xinyu Zhao +7
Generative recommendation models based on autoregressive decoding have been adopted as alternatives to classical ranking-based systems. While their modeling flexibility has been wi…
From Uncertainty to Action: Learning to Steer LLM Agents
Hanwen Li, Jinhao Duan, Guanhua Zhu +4
Steering an LLM agent means deciding whether to correct it, at which step, and with which mechanism. Uncertainty is often used to decide when to correct an agent, but whether it ca…
Test-Time Unlearning via Sparse Autoencoder
Pingzhi Li, Jinhao Duan, Vaishnav Tadiparthi +6
Machine unlearning aims to remove specific knowledge from a trained large language model (LLM) without retraining from scratch. Existing methods modify model weights via gradient a…
"very likely" Means "uncertain"? How LLMs Diverge from Humans in Linguistic Uncertainty Quantification
Jinhao Duan, Zicheng Liu, Zijie Liu +2
Humans express uncertainty verbally via markers (e.g., "possible," "likely"), yet most LLM uncertainty quantification (UQ) relies on costing likelihood- or consistency-based signal…
HarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety
Yajing Bai, Jinhao Duan, Jie Peng +4
Large language models are increasingly deployed through agent harnesses that manage tools, extensions, persistent state, permissions, and external actions. Existing safety benchmar…
Stochastic Meta-Unlearning: Bridging Language Backbone and Multimodal Unlearning
Zijie Liu, Jinhao Duan, Gaowen Liu +2
Machine unlearning for vision-language models (VLMs) remains underexplored. Unlike language models, VLMs combine a language backbone with visual components, which makes unlearning…
End-to-End Dynamic Sparsity for Resource-Adaptive LLM Inference
Yuhang Chen, Jinhao Duan, Ruichen Zhang +11
Large Language Models (LLMs) inference is typically deployed under a static resource assumption, where models execute a fixed computational graph regardless of the runtime environm…
Bifocal Diffusion Language Models: Asymmetric Bidirectional Context for Parallel Generation
Yuhang Chen, Xianfeng Wu, Jinhao Duan +11
Discrete diffusion language models (dLLMs) recover masked tokens in parallel, offering significant speedups over autoregressive (AR) generation. However, such promising frameworks…
SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents
Hao Cheng, Changtao Miao, Tianle Song +21
Autonomous LLM agents increasingly operate in stateful environments where they access tools, files, memory, and external services. While such capabilities enable complex real-world…
IUQ: Interrogative Uncertainty Quantification for Long-Form Large Language Model Generation
Haozhi Fan, Jinhao Duan, Kaidi Xu
Despite the rapid advancement of Large Language Models (LLMs), uncertainty quantification in LLM generation is a persistent challenge. Although recent approaches have achieved stro…
A Replicate-and-Quantize Strategy for Plug-and-Play Load Balancing of Sparse Mixture-of-Experts LLMs
Zijie Liu, Jie Peng, Jinhao Duan +7
Sparse Mixture-of-Experts (SMoE) architectures are increasingly used to scale large language models efficiently, delivering strong accuracy under fixed compute budgets. However, SM…
EComStage: Stage-wise and Orientation-specific Benchmarking for Large Language Models in E-commerce
Kaiyan Zhao, Zijie Meng, Zheyong Xie +4
Large Language Model (LLM)-based agents are increasingly deployed in e-commerce applications to assist customer services in tasks such as product inquiries, recommendations, and or…
The Imitation Game: Using Large Language Models as Chatbots to Combat Chat-Based Cybercrimes
Yifan Yao, Baojuan Wang, Jinhao Duan +4
Chat-based cybercrime has emerged as a pervasive threat, with attackers leveraging real-time messaging platforms to conduct scams that rely on trust-building, deception, and psycho…
Conformal Lesion Segmentation for 3D Medical Images
Binyu Tan, Zhiyuan Wang, Jinhao Duan +4
Medical image segmentation serves as a critical component of precision medicine, enabling accurate localization and delineation of pathological regions, such as lesions. However, e…
Sparse Neurons Carry Strong Signals of Question Ambiguity in LLMs
Zhuoxuan Zhang, Jinhao Duan, Edward Kim +1
Ambiguity is pervasive in real-world questions, yet large language models (LLMs) often respond with confident answers rather than seeking clarification. In this work, we show that…
LRR-Bench: Left, Right or Rotate? Vision-Language models Still Struggle With Spatial Understanding Tasks
Fei Kong, Jinhao Duan, Kaidi Xu +3
Real-world applications, such as autonomous driving and humanoid robot manipulation, require precise spatial perception. However, it remains underexplored how Vision-Language Model…
COIN: Uncertainty-Guarding Selective Question Answering for Foundation Models with Provable Risk Guarantees
Zhiyuan Wang, Jinhao Duan, Qingni Wang +4
Uncertainty quantification (UQ) for foundation models is essential to identify and mitigate potential hallucinations in automatically generated text. However, heuristic UQ approach…
UProp: Investigating the Uncertainty Propagation of LLMs in Multi-Step Agentic Decision-Making
Jinhao Duan, James Diffenderfer, Sandeep Madireddy +3
As Large Language Models (LLMs) are integrated into safety-critical applications involving sequential decision-making in the real world, it is essential to know when to trust LLM d…
TruthPrInt: Mitigating Large Vision-Language Models Object Hallucination Via Latent Truthful-Guided Pre-Intervention
Jinhao Duan, Fei Kong, Hao Cheng +6
Object Hallucination (OH) has been acknowledged as one of the major trustworthy challenges in Large Vision-Language Models (LVLMs). Recent advancements in Large Language Models (LL…
DynaCode: A Dynamic Complexity-Aware Code Benchmark for Evaluating Large Language Models in Code Generation
Wenhao Hu, Jinhao Duan, Chunchen Wei +3
The rapid advancement of large language models (LLMs) has significantly improved their performance in code generation tasks. However, existing code benchmarks remain static, consis…
GuideLLM: Exploring LLM-Guided Conversation with Applications in Autobiography Interviewing
Jinhao Duan, Xinyu Zhao, Zhuoxuan Zhang +12
Although Large Language Models (LLMs) succeed in human-guided conversations such as instruction following and question answering, the potential of LLM-guided conversations-where LL…
Jailbreak-AudioBench: In-Depth Evaluation and Analysis of Jailbreak Threats for Large Audio Language Models
Hao Cheng, Erjia Xiao, Jing Shao +6
Large Language Models (LLMs) demonstrate impressive zero-shot performance across a wide range of natural language processing tasks. Integrating various modality encoders further ex…
ConU: Conformal Uncertainty in Large Language Models with Correctness Coverage Guarantees
Zhiyuan Wang, Jinhao Duan, Lu Cheng +6
Uncertainty quantification (UQ) in natural language generation (NLG) tasks remains an open challenge, exacerbated by the closed-source nature of the latest large language models (L…
Transfer Attack for Bad and Good: Explain and Boost Adversarial Transferability across Multimodal Large Language Models
Hao Cheng, Erjia Xiao, Jiayan Yang +8
Multimodal Large Language Models (MLLMs) demonstrate exceptional performance in cross-modality interaction, yet they also suffer adversarial vulnerabilities. In particular, the tra…