AI研究論文デイリー

MAP-Neo: Highly Capable and Transparent Bilingual Large Language Model Series

Self-Exploring Language Models: Active Preference Elicitation for Online Alignment

T2V-Turbo: Breaking the Quality Bottleneck of Video Consistency Model with Mixed Reward Feedback

LLMs achieve adult human performance on higher-order theory of mind tasks

Offline Regularised Reinforcement Learning for Large Language Models Alignment

Nearest Neighbor Speculative Decoding for LLM Generation and Attribution

EasyAnimate: A High-Performance Long Video Generation Method based on Transformer Architecture

Zipper: A Multi-Tower Decoder Architecture for Fusing Modalities

NPGA: Neural Parametric Gaussian Avatars

Value-Incentivized Preference Optimization: A Unified Approach to Online and Offline RLHF

SoundCTM: Uniting Score-based and Consistency Models for Text-to-Sound Generation

Atlas3D: Physically Constrained Self-Supporting Text-to-3D for Simulation and Fabrication