🔄 In a Training Loop

Urro

urroxyz

98 995 107

https://urro.xyz/

urroxyz

AI & ML interests

computational linguistics major 🤖🔎🔠 i am autistic. if i come off rude, i probably didn't mean to. please feel free to ask me for clarification.

Recent Activity

updated a collection about 9 hours ago

WTF GENIUS PAPERS

updated a collection about 9 hours ago

WTF GENIUS PAPERS

upvoted a paper about 9 hours ago

Sharpen Without Search: On-Policy Distillation of Sequence-Level Power Distribution

View all activity

Organizations

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

Sharpen Without Search: On-Policy Distillation of Sequence-Level Power Distribution

Paper • 2610.06804 • Published 4 days ago • 6

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches

Paper • 2610.06647 • Published 4 days ago • 111

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted 2 papers about 9 hours ago

HuatuoGPT-3: RL-Only Domain Adaptation from Base Models

Paper • 2610.05966 • Published 4 days ago • 29

Labels Override Definitions in Jev-Style Typed Decision Models

Paper • 2610.02586 • Published 8 days ago • 7

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

Stepped MoE: Segment-Level Routing with Configurable Inference Complexity

Paper • 2610.07348 • Published 4 days ago • 4

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

OPD Before RL: Warm-Starting Rubric-Based RL with On-Policy Distillation

Paper • 2610.02781 • Published 7 days ago • 7

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

Multilinguality in Hybrid Attention LLMs

Paper • 2609.35378 • Published 11 days ago • 5

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

UNREAL: Unifying Retrieval and Long-Context with a Single Model

Paper • 2610.08463 • Published 3 days ago • 21

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

HLA: Expressive Hybrid Linear Attention via Chunk-Wise Dynamic Mixing

Paper • 2610.05842 • Published 4 days ago • 7

updated a collection about 9 hours ago

WTF GENIUS PAPERS

Collection

Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 444 items • Updated about 9 hours ago • 98

upvoted a paper about 9 hours ago

Learning Functional Subspaces for Neural Network Compression

Paper • 2609.40127 • Published 9 days ago • 5

Urro

AI & ML interests

Recent Activity

Organizations

urroxyz's activity