5 citations · 5 across the 6 of their papers we have counts for
8 papers
Video Evidence Indexing: Learning Where to Look from Video Previews for Token-Budgeted Long-Video Question Answering
Haowen Guan, Shengzhi Li, Shichao Pei
Long-video question answering is limited by the high cost of visual tokens and by the fixed context width of current VLMs. A long-video question may require broad temporal coverage…
AppliedScientist: Automated Scientific Revision Through Iterative AI Reviewing
Vidushee Vats, Karun Sharma, Shengzhi Li +1
Automated reviewing systems are increasingly evaluated based on the quality of the reviews they produce. Yet a review is only useful if acting on it leads to a measurable improveme…
Predictive Prefetching for Retrieval-Augmented Generation
Wuyang Zhang, Shichao Pei
Retrieval-Augmented Generation (RAG) improves factual grounding in large language models but suffers from substantial latency due to synchronous retrieval. While recent work explor…
PushupBench: Your VLM is not good at counting pushups
Shengzhi Li, Jiarun Chen, Karun Sharma +2
Large vision-language models (VLMs) can recognize \textit{what} happens in video but fail to count \textit{how many} times. We introduce \textbf{PushupBench}, 446 long-form clips (…
Behavioral Consistency and Transparency Analysis on Large Language Model API Gateways
Guanjie Lin, Yinxin Wan, Shichao Pei +3
Third-party Large Language Model (LLM) API gateways are rapidly emerging as unified access points to models offered by multiple vendors. However, the internal routing, caching, and…
Your LLM Agent Can Leak Your Data: Data Exfiltration via Backdoored Tool Use
Wuyang Zhang, Shichao Pei
Tool-use large language model (LLM) agents are increasingly deployed to support sensitive workflows, relying on tool calls for retrieval, external API access, and session memory ma…
Abstract2Appendix: Academic Reviews Enhance LLM Long-Context Capabilities
Shengzhi Li, Kittipat Kampa, Rongyu Lin +2
Large language models (LLMs) have shown remarkable performance across various tasks, yet their ability to handle long-context reading remains challenging. This study explores the e…
Multi-modal Preference Alignment Remedies Degradation of Visual Instruction Tuning on Language Models
Shengzhi Li, Rongyu Lin, Shichao Pei
Multi-modal large language models (MLLMs) are expected to support multi-turn queries of interchanging image and text modalities in production. However, the current MLLMs trained wi…