AI 연구 논문 데일리

Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling

SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators

Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning

Training Language Models for Social Deduction with Multi-Agent Reinforcement Learning

The Curse of Depth in Large Language Models

LM2: Large Memory Models

Matryoshka Quantization

CODESIM: Multi-Agent Code Generation and Problem Solving through Simulation-Driven Planning and Debugging

Show-o Turbo: Towards Accelerated Unified Multimodal Understanding and Generation

ReasonFlux: Hierarchical LLM Reasoning via Scaling Thought Templates

MetaChain: A Fully-Automated and Zero-Code Framework for LLM Agents

Lossless Acceleration of Large Language Models with Hierarchical Drafting based on Temporal Locality in Speculative Decoding

Lumina-Video: Efficient and Flexible Video Generation with Multi-scale Next-DiT

EVEv2: Improved Baselines for Encoder-Free Vision-Language Models

The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information Steering

History-Guided Video Diffusion

CustomVideoX: 3D Reference Attention Driven Dynamic Adaptation for Zero-Shot Customized Video Diffusion Transformers

Dual Caption Preference Optimization for Diffusion Models

Efficient-vDiT: Efficient Video Diffusion Transformers With Attention Tile

DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference Optimization

Towards Internet-Scale Training For Agents

APE: Faster and Longer Context-Augmented Generation via Adaptive Parallel Encoding

Jakiro: Boosting Speculative Decoding with Decoupled Multi-Head via MoE

Steel-LLM:From Scratch to Open Source -- A Personal Journey in Building a Chinese-Centric LLM

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests

Embodied Red Teaming for Auditing Robotic Foundation Models