7 papers
When Models Hear What They Expect: Diagnosing Prosodic Heuristics in Multimodal Sarcasm Detection
Yongjian Chen, Pengfei Wei, Yiqun Sun +2
Multimodal Large Language Models (MLLMs) process speech and text jointly, yet whether they exploit prosodic cues for pragmatic inference or rely on surface acoustic patterns has re…
GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models
Yiqun Sun, Junyu Chen, Pengfei Wei +1
Generative vision-language models (VLMs) are increasingly used in human-centered settings, yet they can produce demographically biased outputs even when images differ only in contr…
Return of Frustratingly Easy Unsupervised Video Domain Adaptation
Pengfei Wei, Yiqun Sun, Zhiqiang Xu +2
Unsupervised video domain adaptation (UVDA) is a practical but under-explored problem. In this paper, we propose a frustratingly easy UVDA method, called MetaTrans. Specifically, M…
Very Efficient Listwise Multimodal Reranking for Long Documents
Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh
Listwise reranking is a key yet computationally expensive component in vision-centric retrieval and multimodal retrieval-augmented generation (M-RAG) over long documents. While rec…
Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs
Hao Wang, Yiqun Sun, Pengfei Wei +2
Vision-language models (VLMs) have advanced rapidly and are increasingly deployed in real-world applications, especially with the rise of agent-based systems. However, their safety…
Corpus2Skill: Distilling Enterprise Knowledge into Navigable Agent Skills for QA and RAG
Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh
Retrieval-Augmented Generation (RAG) grounds LLM responses in external evidence but treats the model as a passive consumer of search results, with no view of how the corpus is orga…
OrchMAS: Orchestrated Reasoning with Multi Collaborative Heterogeneous Scientific Expert Structured Agents
Yichao Feng, Haoran Luo, Zhenghong Lin +4
Multi-agent large language model frameworks are promising for complex multi step reasoning, yet existing systems remain weak for scientific and knowledge intensive domains due to s…