活動海報與簡報

Learnings from using tens of neurons to billions 

Self-Augmented Preference Alignment for Sycophancy Reduction in LLMs

Confidence-Driven Multi-Scale Model Selection for Cost-Efficient Inference

Behavior-Aware Conversational Agents: From Strategy Injection to Personalized Interaction Planning

SMUTF: Schema Matching Using Generative Tags and Hybrid Features

VQA-Induct: Instruction Induction for Visual Question Answering   

SR for Affective Speech: Investigating Impact of Emotion and Speech Generative Strategy

Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs

RE-LLM: Refining Empathetic Speech-LLM Responses by Integrating Emotion Nuance

Positional Cognitive Specialization: Where Do LLMs Learn to Comprehend and Speak Your Language?

Multimodal Knowledge Graph-Based Retrieval Augmented Generation