Models
Datasets
Spaces
Buckets new
Docs
Enterprise
Pricing
- Website
- Community
- Solutions
Log In
Sign Up

Collections

Discover the best community collections!

Collections including paper arxiv:2605.18643

Reinforcement learning

Flash-GRPO: Efficient Alignment for Video Diffusion via One-Step Policy Optimization

Paper • 2605.15980 • Published May 15 • 36
NGRPO: Negative-enhanced Group Relative Policy Optimization

Paper • 2509.18851 • Published Sep 23, 2025 • 2
CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization

Paper • 2605.19436 • Published about 1 month ago • 14
Delta Attention Residuals

Paper • 2605.18855 • Published May 13 • 8

ShotStream: Streaming Multi-Shot Video Generation for Interactive Storytelling

Paper • 2603.25746 • Published Mar 26 • 155
TAPS: Task Aware Proposal Distributions for Speculative Sampling

Paper • 2603.27027 • Published Mar 27 • 144
Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models

Paper • 2603.25716 • Published Mar 26 • 156
LongCat-Next: Lexicalizing Modalities as Discrete Tokens

Paper • 2603.27538 • Published Mar 29 • 147

Agentic AI Training and Tuning

Tongyi DeepResearch Technical Report

Paper • 2510.24701 • Published Oct 28, 2025 • 104
Kimi Linear: An Expressive, Efficient Attention Architecture

Paper • 2510.26692 • Published Oct 30, 2025 • 133
Natural-Language Agent Harnesses

Paper • 2603.25723 • Published Mar 26 • 27
CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery

Paper • 2604.01658 • Published Apr 2 • 55

EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters

Paper • 2402.04252 • Published Feb 6, 2024 • 31
Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models

Paper • 2402.03749 • Published Feb 6, 2024 • 15
ScreenAI: A Vision-Language Model for UI and Infographics Understanding

Paper • 2402.04615 • Published Feb 7, 2024 • 45
EfficientViT-SAM: Accelerated Segment Anything Model Without Performance Loss

Paper • 2402.05008 • Published Feb 7, 2024 • 24

about 1 month ago

TsinghuaC3I/ZEDA-Qwen3-30B-A3B-Dynamic

Text Generation • 31B • Updated 28 days ago • 39 • 1
TsinghuaC3I/ZEDA-GLM-4.7-Flash-Dynamic

Text Generation • 30B • Updated 28 days ago • 124 • 2
TsinghuaC3I/ZEDA

Preview • Updated 28 days ago • 120 • 1
Post-Trained MoE Can Skip Half Experts via Self-Distillation

Paper • 2605.18643 • Published May 18 • 30

LLM Architectures

about 3 hours ago

Kimi Linear: An Expressive, Efficient Attention Architecture

Paper • 2510.26692 • Published Oct 30, 2025 • 133
GLM-5: from Vibe Coding to Agentic Engineering

Paper • 2602.15763 • Published Feb 17 • 164
Believe Your Model: Distribution-Guided Confidence Calibration

Paper • 2603.03872 • Published Mar 4 • 40
OpenWorldLib: A Unified Codebase and Definition of Advanced World Models

Paper • 2604.04707 • Published Apr 6 • 203

Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free

Paper • 2410.10814 • Published Oct 14, 2024 • 51
Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment

Paper • 2502.16894 • Published Feb 24, 2025 • 33
Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs

Paper • 2506.14731 • Published Jun 17, 2025 • 8
SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation

Paper • 2506.18349 • Published Jun 23, 2025 • 13

Reinforcement learning

Flash-GRPO: Efficient Alignment for Video Diffusion via One-Step Policy Optimization

Paper • 2605.15980 • Published May 15 • 36
NGRPO: Negative-enhanced Group Relative Policy Optimization

Paper • 2509.18851 • Published Sep 23, 2025 • 2
CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization

Paper • 2605.19436 • Published about 1 month ago • 14
Delta Attention Residuals

Paper • 2605.18855 • Published May 13 • 8

about 1 month ago

TsinghuaC3I/ZEDA-Qwen3-30B-A3B-Dynamic

Text Generation • 31B • Updated 28 days ago • 39 • 1
TsinghuaC3I/ZEDA-GLM-4.7-Flash-Dynamic

Text Generation • 30B • Updated 28 days ago • 124 • 2
TsinghuaC3I/ZEDA

Preview • Updated 28 days ago • 120 • 1
Post-Trained MoE Can Skip Half Experts via Self-Distillation

Paper • 2605.18643 • Published May 18 • 30

ShotStream: Streaming Multi-Shot Video Generation for Interactive Storytelling

Paper • 2603.25746 • Published Mar 26 • 155
TAPS: Task Aware Proposal Distributions for Speculative Sampling

Paper • 2603.27027 • Published Mar 27 • 144
Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models

Paper • 2603.25716 • Published Mar 26 • 156
LongCat-Next: Lexicalizing Modalities as Discrete Tokens

Paper • 2603.27538 • Published Mar 29 • 147

LLM Architectures

about 3 hours ago

Kimi Linear: An Expressive, Efficient Attention Architecture

Paper • 2510.26692 • Published Oct 30, 2025 • 133
GLM-5: from Vibe Coding to Agentic Engineering

Paper • 2602.15763 • Published Feb 17 • 164
Believe Your Model: Distribution-Guided Confidence Calibration

Paper • 2603.03872 • Published Mar 4 • 40
OpenWorldLib: A Unified Codebase and Definition of Advanced World Models

Paper • 2604.04707 • Published Apr 6 • 203

Agentic AI Training and Tuning

Tongyi DeepResearch Technical Report

Paper • 2510.24701 • Published Oct 28, 2025 • 104
Kimi Linear: An Expressive, Efficient Attention Architecture

Paper • 2510.26692 • Published Oct 30, 2025 • 133
Natural-Language Agent Harnesses

Paper • 2603.25723 • Published Mar 26 • 27
CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery

Paper • 2604.01658 • Published Apr 2 • 55

Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free

Paper • 2410.10814 • Published Oct 14, 2024 • 51
Make LoRA Great Again: Boosting LoRA with Adaptive Singular Values and Mixture-of-Experts Optimization Alignment

Paper • 2502.16894 • Published Feb 24, 2025 • 33
Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs

Paper • 2506.14731 • Published Jun 17, 2025 • 8
SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation

Paper • 2506.18349 • Published Jun 23, 2025 • 13

EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters

Paper • 2402.04252 • Published Feb 6, 2024 • 31
Vision Superalignment: Weak-to-Strong Generalization for Vision Foundation Models

Paper • 2402.03749 • Published Feb 6, 2024 • 15
ScreenAI: A Vision-Language Model for UI and Infographics Understanding

Paper • 2402.04615 • Published Feb 7, 2024 • 45
EfficientViT-SAM: Accelerated Segment Anything Model Without Performance Loss

Paper • 2402.05008 • Published Feb 7, 2024 • 24

Company

TOS Privacy About Careers

Website

Models Datasets Spaces Pricing Docs