khtsly

khtsly's activity

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

upvoted 2 papers 4 days ago

Delta Attention Residuals

Paper • 2605.18855 • Published May 13 • 9

Recursive Language Models

Paper • 2512.24601 • Published Dec 31, 2025 • 102

upvoted a paper 5 days ago

Multi-Token Prediction Done Right

Paper • 2610.38331 • Published 8 days ago • 22

upvoted a paper 8 days ago

RSI Arena: Towards Recursive Self-Improvement with Human Feedback

Paper • 2610.33624 • Published 12 days ago • 27

upvoted a paper 10 days ago

A Fix for Softmax and Cross-Entropy

Paper • 2610.31642 • Published 14 days ago • 4

upvoted a paper 13 days ago

Cache Engineering for Agents

Paper • 2610.28577 • Published 16 days ago • 104

upvoted a paper 14 days ago

Breaking Diffusion: Parallel Reasoning with Draft Refinement

Paper • 2610.26542 • Published 17 days ago • 24

upvoted a paper 15 days ago

Kronecker Delta Rule for Long Context Linear Attention

Paper • 2610.26834 • Published 17 days ago • 31

upvoted a paper 16 days ago

Virtual Head Attention

Paper • 2610.17959 • Published 23 days ago • 17

upvoted a paper 17 days ago

Native Hybrid Attention for Efficient Sequence Modeling

Paper • 2510.07019 • Published Oct 8, 2025 • 18

upvoted 4 papers 20 days ago

Compressed Convolutional Attention: Efficient Attention in a Compressed Latent Space

Paper • 2510.04476 • Published Oct 6, 2025 • 17

Continuum Attention

Paper • 2610.20586 • Published 21 days ago • 8

Long-Horizon RL Needs Muon

Paper • 2610.20256 • Published 22 days ago • 12

LatentLoop: Scaling Looped MoE Models with Asymmetric Compute Allocation

Paper • 2610.19975 • Published 22 days ago • 44

upvoted 4 papers 21 days ago

UltraBERT: Beyond the Limit of BERT Pre-Training

Paper • 2610.15028 • Published 26 days ago • 62

LeWAM: End-to-End World and Action Modeling with JEPAs

Paper • 2610.12818 • Published 28 days ago • 21

Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes

Paper • 2610.18356 • Published 23 days ago • 118

Jeva: System One Models from First Principles

Paper • 2610.18773 • Published 23 days ago • 140

upvoted a paper 22 days ago

Multi-Stream Attention Residuals

Paper • 2610.13630 • Published 27 days ago • 43

upvoted a paper 23 days ago

RSI-Anything: Unlocking Recursive Self-Improvement with World Models

Paper • 2610.13912 • Published 27 days ago • 38