2025-11-17

Models

• SAM 3
  • Introducing Meta Segment Anything Model 3 and Segment Anything Playground
  • GitHub - facebookresearch/sam3: The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
  • SAM 3: Segment Anything with Concepts | Research - AI at Meta
• Introducing SAM 3D: Powerful 3D Reconstruction for Physical World Images
  • SAM 3D: 3Dfy Anything in Images | Research - AI at Meta
  • SAM 3D Body: Robust Full-Body Human Mesh Recovery | Research - AI at Meta
  • GitHub - facebookresearch/sam-3d-body: The repository provides code for running inference with the SAM 3D Body Model (3DB), links for downloading the trained model checkpoints and datasets, and example notebooks that show how to use the model.
  • GitHub - facebookresearch/sam-3d-objects: SAM 3D Objects
• Olmo 3

Papers

• [2511.11238] Virtual Width Networks
• [2507.17702] Towards Greater Leverage: Scaling Laws for Efficient Mixture-of-Experts Language Models
• [2508.14879] MeshCoder: LLM-Powered Structured Mesh Code Generation from Point Clouds
• [2511.11270] Φeat: Physically-Grounded Feature Representation
• [2511.14582] OmniZip: Audio-Guided Dynamic Token Compression for Fast Omnimodal Large Language Models
• [2511.13720] Back to Basics: Let Denoising Generative Models Denoise
• [2507.06203] A Survey on Latent Reasoning
• [2511.13189] Large Language Models Meet Extreme Multi-label Classification: Scaling and Multi-modal Framework
• [2511.09611] MMaDA-Parallel: Multimodal Large Diffusion Language Models for Thinking-Aware Editing and Generation
• [2511.14210] Orion: A Unified Visual Agent for Multimodal Perception, Advanced Visual Reasoning and Execution
• [2511.13647] Part-X-MLLM: Part-aware 3D Multimodal Large Language Model
• [2511.13648] PhysX-Anything: Simulation-Ready Physical 3D Assets from Single Image
• [2511.13254] Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Performance
• GitHub - yujunwei04/UnSAMv2: Code release for “UnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity”
• [2509.03505] LimiX: Unleashing Structured-Data Modeling Capability for Generalist Intelligence
• [2511.14329] Step by Step Network
• [2511.14716] Diffusion As Self-Distillation: End-to-End Latent Diffusion In One Model
• Evolution Strategies at the Hyperscale
• First Frame Is the Place to Go for Video Content Customization
• [2511.16664] Nemotron Elastic: Towards Efficient Many-in-One Reasoning LLMs

Code

• GitHub - NVIDIA/cuda-checkpoint: CUDA checkpoint and restore utility

Articles

• grpo bois
• Run FLUX.1-dev three times faster | Modal Blog
• Host overhead is killing your inference efficiency | Modal Blog

Videos

• Low bit dtypes, sparsity and determinism - YouTube
• KEYNOTE Ralf Gommers - Big ideas shaping scientific Python: the quest for performance and usability - YouTube

Other

• Code Wiki

Tweets