AI Research Papers Daily

Self-rewarding correction for mathematical reasoning

MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning

R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts

NeoBERT: A Next-Generation BERT

LongRoPE2: Near-Lossless LLM Context Window Scaling

UniTok: A Unified Tokenizer for Visual Generation and Understanding

Multimodal Representation Alignment for Image Generation: Text-Image Interleaved Control Is Easier Than You Think

FINEREASON: Evaluating and Improving LLMs' Deliberate Reasoning through Reflective Puzzle Solving

CODESYNC: Synchronizing Large Language Models with Dynamic Code Evolution at Scale

FlexiDiT: Your Diffusion Transformer Can Easily Generate High-Quality Samples with Less Compute

Mobius: Text to Seamless Looping Video Generation via Latent Shift

Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation

Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting

Lean and Mean: Decoupled Value Policy Optimization with Global Value Guidance

Guardians of the Agentic System: Preventing Many Shots Jailbreak with Agentic System

SoRFT: Issue Resolving with Subtask-oriented Reinforced Fine-Tuning

PlanGEN: A Multi-Agent Framework for Generating Planning and Reasoning Trajectories for Complex Problem Solving

R1-T1: Fully Incentivizing Translation Capability in LLMs via Reasoning Learning

Training Consistency Models with Variational Noise Coupling

On Relation-Specific Neurons in Large Language Models

Efficient Gaussian Splatting for Monocular Dynamic Scene Rendering via Sparse Time-Variant Attribute Modeling