Shaobai Jiang

shaobaij's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 8 minutes ago

FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention

Paper • 2606.09079 • Published 22 days ago • 65

upvoted a paper about 2 hours ago

Autodata: An agentic data scientist to create high quality synthetic data

Paper • 2606.25996 • Published 6 days ago • 16

upvoted a paper about 4 hours ago

FLAT: Feedforward Latent Triangle Splatting for Geometrically Accurate Scene Generation

Paper • 2606.24876 • Published 7 days ago • 22

upvoted 2 papers about 5 hours ago

Epicure: Navigating the Emergent Geometry of Food Ingredient Embeddings

Paper • 2605.22391 • Published May 21 • 41

Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems

Paper • 2605.26302 • Published May 25 • 33

upvoted 2 papers about 13 hours ago

Are We Ready For An Agent-Native Memory System?

Paper • 2606.24775 • Published 7 days ago • 108

OpenThoughts-Agent: Data Recipes for Agentic Models

Paper • 2606.24855 • Published 7 days ago • 46

upvoted a paper about 14 hours ago

Unlimited OCR Works

Paper • 2606.23050 • Published 8 days ago • 41

upvoted 8 papers 1 day ago

Self-Harness: Harnesses That Improve Themselves

Paper • 2606.09498 • Published 22 days ago • 1

Compiling Agentic Workflows into LLM Weights: Near-Frontier Quality at Two Orders of Magnitude Less Cost

Paper • 2605.22502 • Published May 21 • 1

Stable Audio 3

Paper • 2605.17991 • Published May 18 • 21

OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization

Paper • 2605.17757 • Published May 18 • 66

Polar: Agentic RL on Any Harness at Scale

Paper • 2605.24220 • Published May 22 • 5

Negligible in Size, Significant in Effect: On Scale Vectors in Large Language Models

Paper • 2605.26895 • Published May 26 • 22

Less is More: Early Stopping Rollout for On-Policy Distillation

Paper • 2605.27028 • Published May 26 • 15

Share More, Search Less: Collaborative Parallel Thinking for Efficient Test-Time Scaling

Paper • 2605.27030 • Published May 26 • 32

upvoted 4 papers 2 days ago

STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?

Paper • 2605.06527 • Published May 7 • 47

Steered LLM Activations are Non-Surjective

Paper • 2604.09839 • Published May 7 • 15

Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation

Paper • 2605.11739 • Published May 13 • 60

MixSD: Mixed Contextual Self-Distillation for Knowledge Injection

Paper • 2605.16865 • Published May 16 • 10