6 papers
Online Pandora's Box for Contextual LLM Cascading
Alexandre Belloni, Yan Chen, Yehua Wei
Motivated by Large Language Model (LLM) cascading, we propose an online contextual Pandora's Box model for adaptively querying and selecting LLM APIs. In each period, a decision-ma…
Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality
Junliang Li, Yucheng Wang, Yan Chen +5
Hallucination in large language models (LLMs) during long-form generation remains difficult to address under existing reinforcement learning from human feedback (RLHF) frameworks,…
SELF-EMO: Emotional Self-Evolution from Recognition to Consistent Expression
Shaowei Zhang, Faqiang Qian, Yan Chen +5
Emotion Recognition in Conversation (ERC) has become a fundamental capability for large language models (LLMs) in human-centric interaction. Beyond accurate recognition, coherent e…
ASTRA: An Automated Framework for Strategy Discovery, Retrieval, and Evolution for Jailbreaking LLMs
Xu Liu, Yan Chen, Kan Ling +4
Despite extensive safety alignment, Large Language Models (LLMs) remain vulnerable to jailbreak attacks. However, existing methods generally lack the capability for continuous lear…
CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference
Yulin Zou, Yan Chen, Wenyan Chen +5
Video streaming analytics is a crucial workload for vision-language model serving, but the high cost of multimodal inference limits scalability. Prior systems reduce inference cost…
Semantic Energy: Detecting LLM Hallucination Beyond Entropy
Huan Ma, Jiadong Pan, Jing Liu +7
Large Language Models (LLMs) are being increasingly deployed in real-world applications, but they remain susceptible to hallucinations, which produce fluent yet incorrect responses…