Blog
No more endless PDFs. Discover the core value of the latest top-tier research in one article.
DynaMoE: Dynamic Token-Level Expert Activation with Layer-Wise Adaptive Capacity for Mixture-of-Experts Neural Networks
CharacterFlywheel: Scaling Iterative Improvement of Engaging and Steerable LLMs in Production
RAISE: Requirement-Adaptive Evolutionary Refinement for Training-Free Text-to-Image Alignment
Unifying Language-Action Understanding and Generation for Autonomous Driving
DreamWorld: Unified World Modeling in Video Generation
AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution
Spectral Attention Steering for Prompt Highlighting
LaST-VLA: Thinking in Latent Spatio-Temporal Space for Vision-Language-Action in Autonomous Driving
ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models
Pri4R: Learning World Dynamics for Vision-Language-Action Models with Privileged 4D Representation
Multi-Head Low-Rank Attention
Minimalist Compliance Control
VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection
$π$-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs
Mesh-Pro: Asynchronous Advantage-guided Ranking Preference Optimization for Artist-style Quadrilateral Mesh Generation
Intersubjectivity as a principle determining physical observables and non-classicality
LongRLVR: Long-Context Reinforcement Learning Requires Verifiable Context Rewards
SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?
Beyond Global Similarity: Towards Fine-Grained, Multi-Condition Multimodal Retrieval
Rethinking Camera Choice: An Empirical Study on Fisheye Camera Properties in Robotic Manipulation
NextAds: Towards Next-generation Personalized Video Advertising
Kiwi-Edit: Versatile Video Editing via Instruction and Reference Guidance
FireRed-OCR Technical Report
Sketch2Colab: Sketch-Conditioned Multi-Human Animation via Controllable Flow Distillation
Can AI Agents Agree?
Qwen3-Coder-Next Technical Report
FACE: A Face-based Autoregressive Representation for High-Fidelity and Efficient Mesh Generation
CHIMERA: Compact Synthetic Data for Generalizable LLM Reasoning
Deep Learning for Financial Time Series: A Large-Scale Benchmark of Risk-Adjusted Performance
Efficient RLVR Training via Weighted Mutual Information Data Selection