Urro

urroxyz's activity

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

updated a collection about 6 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. • 54 items • Updated about 6 hours ago • 4

upvoted 2 papers about 20 hours ago

Beyond Fixed Frames: Dynamic Character-Aligned Speech Tokenization

Paper • 2601.23174 • Published 9 days ago • 2

Learning Rate Matters: Vanilla LoRA May Suffice for LLM Fine-tuning

Paper • 2602.04998 • Published 4 days ago • 3

commented a paper about 20 hours ago

Late-to-Early Training: LET LLMs Learn Earlier, So Faster and Better

Paper • 2602.05393 • Published 4 days ago • 5 •

3

updated a collection about 20 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. • 54 items • Updated about 6 hours ago • 4

upvoted 2 papers about 20 hours ago

Late-to-Early Training: LET LLMs Learn Earlier, So Faster and Better

Paper • 2602.05393 • Published 4 days ago • 5

DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers

Paper • 2602.02016 • Published 6 days ago • 9

updated a collection 1 day ago

HUMAN-WRITTEN & LEGALLY-SOURCED*

Collection

Datasets written by humans and/or reverse-engineered from text with deterministic algorithms. No illegal scraping or unethical synthesis. *Mostly. • 155 items • Updated 1 day ago • 2

upvoted a paper 1 day ago

Privileged Information Distillation for Language Models

Paper • 2602.04942 • Published 4 days ago • 20

updated a collection 2 days ago

HUMAN-WRITTEN & LEGALLY-SOURCED*

Collection

Datasets written by humans and/or reverse-engineered from text with deterministic algorithms. No illegal scraping or unethical synthesis. *Mostly. • 155 items • Updated 1 day ago • 2

liked a dataset 2 days ago

uw-math-ai/theorem-search-dataset

Viewer • Updated 3 days ago • 2.89M • 71 • 9

updated a collection 3 days ago

HUMAN-WRITTEN & LEGALLY-SOURCED*

Collection

Datasets written by humans and/or reverse-engineered from text with deterministic algorithms. No illegal scraping or unethical synthesis. *Mostly. • 155 items • Updated 1 day ago • 2

liked a dataset 3 days ago

DataMuncher-Labs/UltiMath

Viewer • Updated 22 days ago • 32.9B • 15.6k • 14

updated a collection 3 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. • 54 items • Updated about 6 hours ago • 4

upvoted a paper 3 days ago

Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probing

Paper • 2602.03845 • Published 5 days ago • 24

updated a collection 3 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. • 54 items • Updated about 6 hours ago • 4

upvoted a paper 3 days ago

Search-R2: Enhancing Search-Integrated Reasoning via Actor-Refiner Collaboration

Paper • 2602.03647 • Published 5 days ago • 7

updated a collection 3 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. • 54 items • Updated about 6 hours ago • 4

upvoted a paper 3 days ago

Horizon-LM: A RAM-Centric Architecture for LLM Training

Paper • 2602.04816 • Published 4 days ago • 16

commented a paper 3 days ago

Chronicals: A High-Performance Framework for LLM Fine-Tuning with 3.51x Speedup over Unsloth

Paper • 2601.02609 • Published Jan 6 • 2 •

2