marinaretikof

marinaretik's activity

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

upvoted an article about 20 hours ago

view article

Article

GLM-5.2: Built for Long-Horizon Tasks

• 

zai-org

•

22 days ago

• 124

upvoted a paper 15 days ago

Qwen-AgentWorld: Language World Models for General Agents

Paper • 2606.24597 • Published 17 days ago • 146

upvoted a collection 15 days ago

Qwen-AgentWorld

Collection

3 items • Updated 15 days ago • 66

upvoted 2 papers 28 days ago

Redesign Mixture-of-Experts Routers with Manifold Power Iteration

Paper • 2606.12397 • Published 30 days ago • 89

Toward Generalist Autonomous Research via Hypothesis-Tree Refinement

Paper • 2606.11926 • Published 30 days ago • 126

upvoted 3 collections 2 months ago

Qwen3-VL

Collection

37 items • Updated Dec 31, 2025 • 752

Gemma 4

Collection

15 items • Updated 29 days ago • 1.02k

Qwen-Scope

Collection

16 items • Updated May 14 • 75

upvoted a paper 2 months ago

Recursive Multi-Agent Systems

Paper • 2604.25917 • Published Apr 28 • 287

upvoted an article 2 months ago

view article

Article

Norm-Preserving Biprojected Abliteration

• 

grimjim

•

Nov 6, 2025

• 87

upvoted 4 papers 3 months ago

DFlash: Block Diffusion for Flash Speculative Decoding

Paper • 2602.06036 • Published Feb 5 • 89

Can Large Language Models Reinvent Foundational Algorithms?

Paper • 2604.05716 • Published Apr 7 • 8

Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning

Paper • 2604.16029 • Published Apr 17 • 23

Elucidating the SNR-t Bias of Diffusion Probabilistic Models

Paper • 2604.16044 • Published Apr 17 • 73

upvoted 5 collections 3 months ago

HLWQ Unified (Weights Q5 + KV Cache Q3)

Collection

Full-stack HLWQ: Q5 weights + torchao INT4 + Q3 KV cache · formerly PolarQuant Unified • 16 items • Updated Apr 18 • 3

HLWQ Models

Collection

Hadamard-Lloyd Weight Quantization · arXiv:2603.29078 · formerly PolarQuant • 26 items • Updated Apr 18 • 1

HLWQ Gemma Models

Collection

Google Gemma family quantized with HLWQ (Hadamard-Lloyd) · formerly PolarQuant Gemma • 5 items • Updated Apr 13 • 5

Gemma 4

Collection

Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. • 36 items • Updated 24 days ago • 229

Qwen3.5-27B HLWQ

Collection

Qwen3.5-27B · HLWQ Q5 weight quantization · formerly PolarQuant • 1 item • Updated Apr 13 • 1

upvoted a paper 3 months ago

MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens

Paper • 2603.23516 • Published Mar 6 • 53