AI研究論文デイリー

Unveiling Encoder-Free Vision-Language Models

FunAudioLLM: Voice Understanding and Generation Foundation Models for Natural Interaction Between Humans and LLMs

AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents

Learning to (Learn at Test Time): RNNs with Expressive Hidden States

RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models

ChartGemma: Visual Instruction-tuning for Chart Reasoning in the Wild

Stark: Social Long-Term Multi-Modal Conversation with Persona Commonsense Knowledge

DotaMath: Decomposition of Thought with Code Assistance and Self-correction for Mathematical Reasoning

LLM-jp: A Cross-organizational Project for the Research and Development of Fully Open Japanese LLMs

Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams

On scalable oversight with weak LLMs judging strong LLMs

Safe Unlearning: A Surprisingly Effective and Generalizable Solution to Defend Against Jailbreak Attacks

HEMM: Holistic Evaluation of Multimodal Foundation Models

CRiM-GS: Continuous Rigid Motion-Aware Gaussian Splatting from Motion Blur Images

Granular Privacy Control for Geolocation with Vision Language Models