Urro

urroxyz's activity

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

upvoted 4 papers about 20 hours ago

LLaDA2.1: Speeding Up Text Diffusion via Token Editing

Paper • 2602.08676 • Published 2 days ago • 53

Echoes as Anchors: Probabilistic Costs and Attention Refocusing in LLM Reasoning

Paper • 2602.06600 • Published 5 days ago • 2

F-GRPO: Don't Let Your Policy Learn the Obvious and Forget the Rare

Paper • 2602.06717 • Published 5 days ago • 68

Aster: Autonomous Scientific Discovery over 20x Faster Than Existing Methods

Paper • 2602.07040 • Published 8 days ago • 1

upvoted 2 papers about 21 hours ago

Judging What We Cannot Solve: A Consequence-Based Approach for Oracle-Free Evaluation of Research-Level Math

Paper • 2602.06291 • Published 5 days ago • 21

Thinking Makes LLM Agents Introverted: How Mandatory Thinking Can Backfire in User-Engaged Agents

Paper • 2602.07796 • Published 3 days ago • 6

upvoted 6 papers 2 days ago

compar:IA: The French Government's LLM arena to collect French-language human prompts and preference data

Paper • 2602.06669 • Published 5 days ago • 6

Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing

Paper • 2602.04837 • Published 7 days ago • 7

QuantLRM: Quantization of Large Reasoning Models via Fine-Tuning Signals

Paper • 2602.02581 • Published 11 days ago • 6

ReMiT: RL-Guided Mid-Training for Iterative LLM Evolution

Paper • 2602.03075 • Published 8 days ago • 5

RaBiT: Residual-Aware Binarization Training for Accurate and Efficient LLMs

Paper • 2602.05367 • Published 6 days ago • 7

Canzona: A Unified, Asynchronous, and Load-Balanced Framework for Distributed Matrix-based Optimizers

Paper • 2602.06079 • Published 7 days ago • 17

upvoted 4 papers 3 days ago

Beyond Fixed Frames: Dynamic Character-Aligned Speech Tokenization

Paper • 2601.23174 • Published 12 days ago • 2

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

Paper • 2602.04998 • Published 7 days ago • 6

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

Paper • 2602.05393 • Published 6 days ago • 7

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

Paper • 2602.02016 • Published 9 days ago • 11

upvoted a paper 4 days ago

Privileged Information Distillation for Language Models

Paper • 2602.04942 • Published 7 days ago • 24

upvoted 3 papers 6 days ago

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

Paper • 2602.03845 • Published 8 days ago • 24

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

Paper • 2602.03647 • Published 8 days ago • 7

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

Paper • 2602.04816 • Published 7 days ago • 16