7 papers
VAUQ: Vision-Aware Uncertainty Quantification for LVLM Self-Evaluation
Seongheon Park, Changdae Oh, Hyeong Kyu Choi +2
Large Vision-Language Models (LVLMs) frequently hallucinate, limiting their safe deployment in real-world applications. Existing LLM self-evaluation methods rely on a model's abili…
Limited Preference Data? Learning Better Reward Model with Latent Space Synthesis
Leitian Tao, Xuefeng Du, Sharon Li
Reward modeling, crucial for aligning large language models (LLMs) with human preferences, is often bottlenecked by the high cost of preference data. Existing textual data synthesi…
Understanding Multimodal LLMs Under Distribution Shifts: An Information-Theoretic Approach
Changdae Oh, Zhen Fang, Shawn Im +2
Multimodal large language models (MLLMs) have shown promising capabilities but struggle under distribution shifts, where evaluation data differ from instruction tuning distribution…
Steer LLM Latents for Hallucination Detection
Seongheon Park, Xuefeng Du, Min-Hsuan Yeh +2
Hallucinations in LLMs pose a significant concern to their safe deployment in real-world applications. Recent approaches have leveraged the latent space of LLMs for hallucination d…
Foundations of Unknown-aware Machine Learning
Xuefeng Du
Ensuring the reliability and safety of machine learning models in open-world deployment is a central challenge in AI safety. This thesis develops both algorithmic and theoretical f…
Challenges and Future Directions of Data-Centric AI Alignment
Min-Hsuan Yeh, Jeffrey Wang, Xuefeng Du +4
As AI systems become increasingly capable and influential, ensuring their alignment with human values, preferences, and goals has become a critical research focus. Current alignmen…