khtsly

khtsly's activity

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

upvoted 3 papers 1 day ago

When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation

Paper • 2609.20511 • Published 3 days ago • 74

DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression

Paper • 2609.19969 • Published 3 days ago • 91

JEPA-Anything: Learning Predictive Models across Different Worlds

Paper • 2609.20800 • Published 3 days ago • 38

upvoted 3 papers 2 days ago

VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention

Paper • 2609.15810 • Published 6 days ago • 35

ProgramDistill: From Interactive Web Apps to Verifiable Reference-Guided SWE Tasks

Paper • 2609.18805 • Published 4 days ago • 47

Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening

Paper • 2609.18708 • Published 4 days ago • 62

upvoted 4 papers 4 days ago

SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking

Paper • 2609.13141 • Published 9 days ago • 62

Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction

Paper • 2609.13285 • Published 12 days ago • 70

Atria Dawn: The Dawn of Agentic Superintelligence

Paper • 2609.15818 • Published 6 days ago • 409

Dream-RSI: Recursive Self-Improvement through Evolving Worlds

Paper • 2609.14858 • Published 6 days ago • 230

upvoted a paper 5 days ago

Language Self-Play For Data-Free Training

Paper • 2509.07414 • Published Sep 9, 2025 • 32

upvoted 2 papers 6 days ago

HyQuant: Hybrid-Precision Quantization for LLM Attention

Paper • 2608.27875 • Published 23 days ago • 30

Solar Open 2 Technical Report

Paper • 2607.20062 • Published Jul 24 • 6

upvoted a paper 7 days ago

Erase-then-Delta Attention: Decoupling Erase and Write Addresses in Delta-Rule Linear Attention

Paper • 2606.26560 • Published Jun 25 • 3

upvoted a paper 11 days ago

Kalman Delta Networks: Uncertainty-aware Associative Memory

Paper • 2609.07816 • Published 13 days ago • 29

upvoted 2 papers 12 days ago

Enoki: Efficient Multi-Level Hallucination Detection

Paper • 2609.00581 • Published 19 days ago • 28

Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation

Paper • 2609.02998 • Published 18 days ago • 17

upvoted 3 papers 13 days ago

RISE: Recursive Improvement via Self-Extrapolating Policy Distillation

Paper • 2609.05295 • Published 16 days ago • 17

Group Adaptive Clipping Policy Optimization

Paper • 2609.00444 • Published 20 days ago • 10

MaxKernel: Agentic Kernel Generation for TPUs

Paper • 2609.04523 • Published 17 days ago • 18