Signal over volume.

SSync: Selective Synergistic Learning for Efficient Video Object-Centric Disc...

SSync: Selective Synergistic Learning for Efficient Video Object-Centric Discovery

FreeStyle: Scalable Dual-Reference Generation via Community LoRA Mining

ImageWAM: Replacing Video Generation with Image Editing for Efficient Robot W...

New Likelihood Framework for Inferring COVID-19 Transmission Dynamics from Co...

Guava: A Universal Harness Framework for Efficient Embodied Manipulation

PAIWorld: Achieving 3D-Consistent Multi-View World Modeling for Robotic Manip...

RNG-Bench: Evaluating MLLMs via Controllable Non-Markov Games for State Tracking

Sumi: The First Large-Scale Open Uniform Diffusion Language Model Pretrained ...

IndustryBench-MIPU: A New Benchmark for Multi-Image Industrial Product Attrib...

XCIENTIST: A Research Harness for Externalizing AI Scientific Reasoning and V...

Turing-RL: Optimizing User Simulators for Indistinguishability via Discrimina...

CEO-BENCH: Evaluating Long-Horizon Strategic Intelligence in AI Agents

LOGOS: A Unified Generative Foundation Model for Multi-Domain Natural Sciences

ACE-EGO-0: Unifying Egocentric Human and Robotic Data for VLA Pretraining

Variable-Width Transformers: Optimizing Capacity via X-Shaped Architectures

UniAR: Unifying Multimodal Understanding and Generation via a Single Discrete...

ActWorld: Bridging the Gap Between Navigation and Object Interaction in Real-...

OPD-Evolver: A Slow-Fast Co-Evolution Framework for Holistic Agent Self-Impro...

TRIAGE: Mitigating Risk Polarization in Medical LLMs via Dialectical Reasoning

ProCUA-SFT: Scaling Computer-Use Agents via 3.1M Synthetic Step-Level Samples

WAPO: Stabilizing RLVR via Positive-Advantage Only Policy Optimization

LoopCoder-v2: Identifying the Optimal Loop Count for Efficient Parallel Loop ...

Classifying by Proxy: An Explainable Ensemble for Sensitive CSAI Detection

SearchGEO Reveals High Vulnerability of LLM Search Agents to Web Content Mani...