Haoyang Su (SII) PRO

Hoyant-Su's activity

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

updated a model 4 days ago

Hoyant-Su/A3-Qwen3-14B

Text Generation • 15B • Updated 4 days ago • 37 • 1

liked a model 4 days ago

Hoyant-Su/A3-Qwen3-14B

Text Generation • 15B • Updated 4 days ago • 37 • 1

published a model 4 days ago

Hoyant-Su/A3-Qwen3-14B

Text Generation • 15B • Updated 4 days ago • 37 • 1

submitted a paper to Daily Papers 14 days ago

JevSpawn: Adaptive Agentic Inference through Compositional Action Spaces

Paper • 2610.00437 • Published 16 days ago • 8

authored a paper 14 days ago

JevSpawn: Adaptive Agentic Inference through Compositional Action Spaces

Paper • 2610.00437 • Published 16 days ago • 8

upvoted a paper 14 days ago

JevSpawn: Adaptive Agentic Inference through Compositional Action Spaces

Paper • 2610.00437 • Published 16 days ago • 8

updated 3 collections about 1 month ago

CLI evaluation: task origin and success criteria

Collection

How do terminal benchmarks construct executable tasks and check outputs, filesystem changes and final container state? • 6 items • Updated Sep 8

Agent context: workspace selection and learned memory

Collection

How do agents select an initial workspace view, compress interaction history and learn what to retain? • 5 items • Updated Sep 8

Agent credit assignment: feedback and supervision

Collection

Which rewards, state comparisons and extra supervision support action-level credit in multi-turn agent training? • 8 items • Updated Sep 8