Signal over volume.

DEMON: Turning Diffusion Models into Real-Time Playable Musical Instruments

GCPO: Enabling Fine-Grained Token Credit Assignment in Discrete Policy Optimi...

GCPO: Enabling Fine-Grained Token Credit Assignment in Discrete Policy Optimization

VisualThink-VLA: Achieving High-Accuracy, Low-Latency Robot Control via Visua...

The Physical AI Inference Gap: Why Faster GPUs Don't Always Mean Faster LLM D...

HL-OutPaint: Achieving High-Resolution, Long-Range Video Outpainting via Coar...

Topology-Aware Handover Framework Achieves 99.8% Success in Multi-UAV Vehicle...

SwanBench-Speech: A New Multi-Dimensional Benchmark for Long-Form Speech Gene...

Function2Scene: Designing 3D Indoor Layouts Based on Human Activity and Needs

Representation Forcing: Eliminating VAE Bottlenecks in Unified Multimodal Models

SAAS: Mitigating Over-Search in Agentic LLMs via Self-Aware Reinforcement Lea...

FLAT-PACK BENCH: Exposing the Spatio-Temporal Reasoning Gaps in Large Vision-...

GDSD: Bypassing Likelihood Bias in Diffusion LLM Reinforcement Learning

Suffix-Anchored Confidence Modulation: Fixing Premature Decoding in Diffusion...

GGT-100K: Scaling Real-World Image Restoration via Generative Ground Truth

LongDS-Bench: Exposing the State-Management Bottleneck in Long-Horizon Data A...

StateKV: Achieving Linear-Time Scaling for Long-Video Understanding in VLMs

ClawTrojan: Uncovering Persistent Multi-Step Trojan Attacks in Agentic Worksp...

Not All Disagreement Is Learnable: Identifying High-Value Tokens in On-Policy...

SANA-Streaming: Achieving Real-Time, High-Resolution Streaming Video Editing ...

MAAT: Solving the Causal Knowledge Gap in LLM Unlearning

TaskMem: Learning What to Memorize via Reinforcement Learning for Multimodal ...

AnyMo: Scaling Any-Modality Human Motion Generation with Large-Scale Multimod...

SwanSphere: Achieving Low-Latency, High-Fidelity Streaming Spatial Audio Gene...

Count Anything: A Generalist Model for Cross-Domain Text-Guided Object Counting