AI 연구 논문 데일리

Parallelized Autoregressive Visual Generation

Offline Reinforcement Learning for LLM Multi-Step Reasoning

CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up

SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation

Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesis

Toward Robust Hyper-Detailed Image Captioning: A Multiagent Approach and Dual Evaluation Metrics for Factuality and Coverage

MixLLM: LLM Quantization with Global Mixed-precision between Output-features and Highly-efficient System Design

TRecViT: A Recurrent Video Transformer

Sequence Matters: Harnessing Video Models in 3D Super-Resolution

Multi-LLM Text Summarization

IDOL: Instant Photorealistic 3D Human Creation from a Single Image

LLMs Lost in Translation: M-ALERT uncovers Cross-Linguistic Safety Gaps

Fietje: An open, efficient LLM for Dutch