Zehao Jin

Lunamos's activity

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

upvoted 2 papers about 2 hours ago

Persona Dosing: Calibrated Activation Steering for Graded Trait Control

Paper • 2609.36388 • Published 5 days ago • 2

Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It

Paper • 2609.36585 • Published 4 days ago • 2

upvoted 3 papers 26 days ago

S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?

Paper • 2608.31100 • Published Aug 31 • 114

Aspire: Can Models Self-Evolve from Vague Goals?

Paper • 2608.31111 • Published Aug 31 • 241

HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?

Paper • 2609.01437 • Published Sep 1 • 294

upvoted a collection 4 months ago

FLAS

Collection

Flow-based Activation Steering • 9 items • Updated Jun 4 • 1

upvoted a paper 4 months ago

How Far Are VLMs from Privacy Awareness in the Physical World? An Empirical Study

Paper • 2605.05340 • Published May 8 • 2

upvoted a paper 5 months ago

Beyond Steering Vector: Flow-based Activation Steering for Inference-Time Intervention

Paper • 2605.05892 • Published May 7 • 1