Basit mustafa

BasitMustafa

·

AI & ML interests

None yet

Recent Activity

liked a model 9 minutes ago

InternScience/Sage-A1.5

liked a model about 8 hours ago

baselabs/mammoth-1

liked a model about 8 hours ago

baselabs/elephant-1

View all activity

Organizations

upvoted an article about 9 hours ago

Article

Multimodal open d1 decision models for the edge

LiquidAI

•

15 days ago

• 42

upvoted a collection about 11 hours ago

EmbeddingGemma

7 items • Updated Sep 4, 2025 • 5

upvoted a collection about 13 hours ago

EgoXperience

EgoXperience: EgoX-Data-6M, EgoX-Bench, and the EgoX model. • 4 items • Updated 2 days ago • 3

upvoted 2 papers 1 day ago

BizBench: Unveiling Hidden Challenges in Deploying Long-Horizon Agents for Real-World Business

Paper • 2610.20424 • Published 8 days ago • 493

MemT-1.0: Towards Memory-Native Evolution

Paper • 2610.26825 • Published 3 days ago • 162

upvoted a collection 1 day ago

MemT-1

MemT-1.0: Towards Memory-Native Evolution • 3 items • Updated 2 days ago • 25

upvoted a paper 2 days ago

Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement Learning

Paper • 2610.23181 • Published 6 days ago • 53

upvoted a collection 3 days ago

SlideOn & mSlideOn

A collection of pretrained sparse autoencoders (SAEs) and fine-tuned sparse retrievers for English and multilingual, long-context, code retrieval • 4 items • Updated 3 days ago • 6

upvoted 2 articles 5 days ago

Article

Score Function Centering: Your Stale Rollout Does Not Have to Be Importance Sampled and Stabilize Your Async RL Training with Trainer-Generator KL

NormalUhr

•

5 days ago

• 2

Article

The Most Intuitive Way to Understand Score Function Centering: Understand Trainer-Generator Mismatch in Async RL from the Score Function

NormalUhr

•

6 days ago

• 9

upvoted a paper 5 days ago

OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding

Paper • 2610.18829 • Published 9 days ago • 62

upvoted a collection 6 days ago

Muse Spark Open

Muse Spark Open model. A 816B parameter model with 42B Active parameters. BF16 and NVFP4 Weights. • 2 items • Updated 7 days ago • 164

upvoted a paper 7 days ago

VEPA: Verified Evolving Formal Policy for Lifelong LLM Agent Guardrails

Paper • 2610.12813 • Published 14 days ago • 2

upvoted a collection 9 days ago

QUEEN (Chess) models

Our models trained to play chess (4B) • 2 items • Updated 19 days ago • 2

upvoted 6 papers 10 days ago

Coding-Agent Benchmarks Should Match Their Users' Task Flows

Paper • 2610.09633 • Published 15 days ago • 15

Video-RSI: Recursive Self-Improvement of Video Understanding Agents via Harness Evolution

Paper • 2609.37950 • Published 23 days ago • 16

Surfing the Activation Wave: Analyzing the Information and Geometry of LLM Hidden States for Speculative Decoding

Paper • 2610.12939 • Published 13 days ago • 18

From Given to Gathered Evidence: Agentic Learning for Longitudinal Medical Reasoning

Paper • 2609.39566 • Published 22 days ago • 19

DataVista: Diagnosing Multimodal LLMs on Data Video Understanding

Paper • 2610.11993 • Published 14 days ago • 18

GraphDecisionBench: Benchmarking System One Models on Graph Tasks

Paper • 2610.06354 • Published 13 days ago • 20