🔄 In a Training Loop

Urro

urroxyz

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

upvoted a paper about 16 hours ago

Learning from Your Own Mistakes: Constructing Learnable Micro-Reflective Trajectories for Self-Distillation

updated a collection about 24 hours ago

WTF GENIUS PAPERS

updated a collection about 24 hours ago

WTF GENIUS PAPERS

View all activity

Organizations

upvoted a paper about 16 hours ago

Learning from Your Own Mistakes: Constructing Learnable Micro-Reflective Trajectories for Self-Distillation

Paper • 2606.18844 • Published 7 days ago • 14

updated a collection about 24 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. • 176 items • Updated about 24 hours ago • 41

upvoted a paper about 24 hours ago

Multi-Turn Reflective Masking Elicits Reasoning in Mask Diffusion Models

Paper • 2606.16700 • Published 9 days ago • 11

updated a collection 1 day 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. • 176 items • Updated about 24 hours ago • 41

upvoted 2 papers 1 day ago

RepSelect: Robust LLM Unlearning via Representation Selectivity

Paper • 2606.17168 • Published 9 days ago • 4

Variable-Width Transformers

Paper • 2606.18246 • Published 8 days ago • 13

updated a collection 1 day 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. • 176 items • Updated about 24 hours ago • 41

upvoted a paper 1 day ago

Rethinking the Role of Efficient Attention in Hybrid Architectures

Paper • 2606.15378 • Published 11 days ago • 17

commented a paper 1 day ago

Morpheus: A Morphology-Aware Neural Tokenizer and Word Embedder for Turkish

Paper • 2606.18717 • Published 7 days ago • 6 •

upvoted a paper 1 day ago

Morpheus: A Morphology-Aware Neural Tokenizer and Word Embedder for Turkish

Paper • 2606.18717 • Published 7 days ago • 6

updated a collection 1 day 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. • 176 items • Updated about 24 hours ago • 41

upvoted 3 papers 1 day ago

STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability

Paper • 2606.19236 • Published 7 days ago • 12

Sumi: Open Uniform Diffusion Language Model from Scratch

Paper • 2606.19005 • Published 7 days ago • 11

The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL

Paper • 2606.19162 • Published 7 days ago • 20

upvoted 2 papers 6 days ago

Learning from the Self-future: On-policy Self-distillation for dLLMs

Paper • 2606.18195 • Published 8 days ago • 74

Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients

Paper • 2606.18216 • Published 8 days ago • 60

commented a paper 6 days ago

LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling

Paper • 2606.18023 • Published 8 days ago • 203 •

updated a collection 6 days 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. • 176 items • Updated about 24 hours ago • 41

upvoted a paper 6 days ago

LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling

Paper • 2606.18023 • Published 8 days ago • 203

Urro

AI & ML interests

Recent Activity

Organizations

urroxyz's activity