Blog
告别冗长 PDF，一文读懂最新顶会与核心期刊的创新价值。
DeepResearch-9K: A Challenging Benchmark Dataset of Deep-Research Agent
ShareVerse: Multi-Agent Consistent Video Generation for Shared World Modeling
How to Peel with a Knife: Aligning Fine-Grained Manipulation with Human Preference
Code2Math: Can Your Code Agent Effectively Evolve Math Problems Through Exploration?
GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation
NeuroSkill(tm): Proactive Real-Time Agentic System Capable of Modeling Human State of Mind
ITO: Images and Texts as One via Synergizing Multiple Alignment and Training-Time Fusion
MIBURI: Towards Expressive Interactive Gesture Synthesis
Real Money, Fake Models: Deceptive Model Claims in Shadow APIs
Preconditioned Score and Flow Matching
SimRecon: SimReady Compositional Scene Reconstruction from Real Videos
EvoSkill: Automated Skill Discovery for Multi-Agent Systems
AMemGym: Interactive Memory Benchmarking for Assistants in Long-Horizon Conversations
Neural Discrimination-Prompted Transformers for Efficient UHD Image Restoration and Enhancement
DualSentinel: A Lightweight Framework for Detecting Targeted Attacks in Black-box LLM via Dual Entropy Lull Pattern
APAO: Adaptive Prefix-Aware Optimization for Generative Recommendation
Recursive Think-Answer Process for LLMs and VLMs
pySpatial: Generating 3D Visual Programs for Zero-Shot Spatial Reasoning
Learning to Draft: Adaptive Speculative Decoding with Reinforcement Learning
How Controllable Are Large Language Models? A Unified Evaluation across Behavioral Granularities
FREE-Edit: Using Editing-aware Injection in Rectified Flow Models for Zero-shot Image-Driven Video Editing
Super Research: Answering Highly Complex Questions with Large Language Models through Super Deep and Super Wide Research
MAC: A Conversion Rate Prediction Benchmark Featuring Labels Under Multiple Attribution Mechanisms
Tether: Autonomous Functional Play with Correspondence-Driven Trajectory Warping
Beyond Language Modeling: An Exploration of Multimodal Pretraining
Utonia: Toward One Encoder for All Point Clouds
ZipMap: Linear-Time Stateful 3D Reconstruction with Test-Time Training
Helios: Real Real-Time Long Video Generation Model
Phi-4-reasoning-vision-15B Technical Report
A Selection Aware View of Black Hole-Galaxy Coevolution at High Redshift