Youssef Amrani

youssefam's activity

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

upvoted 2 papers about 10 hours ago

QuantWM: Temporally Consistent 2-Bit KV Cache Quantization for Video World Models

Paper • 2609.26425 • Published 10 days ago • 18

Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory

Paper • 2610.02521 • Published 7 days ago • 57

upvoted a paper 2 days ago

EgoTools: Towards Tool-Centric Reasoning in Real-World Egocentric Videos

Paper • 2609.39378 • Published 8 days ago • 70

upvoted a paper 3 days ago

It's Not What the Image Shows: Irrelevant Context Destabilises VLM Judges Without Informing Them

Paper • 2609.37863 • Published 9 days ago • 38

upvoted 2 papers 11 days ago

StudentSim: Training LLM-based Student Simulators

Paper • 2609.01591 • Published Sep 1 • 496

The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks

Paper • 2609.25804 • Published 16 days ago • 164

upvoted a paper 14 days ago

Harness-Zero: Harness Distillation via Agent-as-Harness

Paper • 2609.24974 • Published 17 days ago • 38

upvoted 3 papers 16 days ago

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

Paper • 2609.19969 • Published 21 days ago • 225

Fathom: Per-Query Read Depth for Sparse Decoding over Offloaded KV Caches

Paper • 2609.17652 • Published 23 days ago • 8

ScienceIDE: Turning World's Scientific Codebase into Agent Learnable Environments

Paper • 2609.19134 • Published 22 days ago • 101

upvoted a paper 17 days ago

Continual Learning Mechanisms Compose for Long-Horizon Memorization

Paper • 2609.06986 • Published Sep 7 • 376

upvoted 2 papers 19 days ago

How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus

Paper • 2609.15504 • Published 24 days ago • 37

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

Paper • 2609.13285 • Published about 1 month ago • 82