AI at Meta

AI at Meta on Hugging Face Daily Papers: 70 papers, 8 in the top 3 of their day, 3 paper of the day.

  1. Scaling Laws for Looped Mixture of Experts 22 upvotes, #30 of 2026-10-01
  2. MoE-ViE: Mixture of Experts Vision Encoder for Efficient Image and Video Understanding 16 upvotes, #13 of 2026-08-19
  3. Skaling: Chinchilla's Exponents Meet Kaplan's Coupling 9 upvotes, #20 of 2026-08-10
  4. Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes 59 upvotes, #3 of 2026-08-06
  5. Reinforcement Learning for Code Optimization 12 upvotes, #15 of 2026-07-29
  6. Two-Level Meta-Rubrics for Evaluating Open-Ended Generation: GAMUT, a Benchmark for Factual Completeness 9 upvotes, #13 of 2026-07-22
  7. Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity 12 upvotes, #8 of 2026-07-09
  8. TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents 47 upvotes, #6 of 2026-06-30
  9. Discretizing Reward Models 17 upvotes, #12 of 2026-06-26
  10. Autodata: An agentic data scientist to create high quality synthetic data 18 upvotes, #11 of 2026-06-25
  11. The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL 20 upvotes, #7 of 2026-06-18
  12. RepFusion: Leveraging Multimodal Priors for Denoising in Representation Space 18 upvotes, #15 of 2026-06-15
  13. MobileMoE: Scaling On-Device Mixture of Experts 14 upvotes, #23 of 2026-05-27
  14. Realiz3D: 3D Generation Made Photorealistic via Domain-Aware Learning 25 upvotes, #16 of 2026-05-15
  15. Sapiens2 16 upvotes, #8 of 2026-04-28
  16. Scaling Test-Time Compute for Agentic Coding 10 upvotes, #14 of 2026-04-23
  17. Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning 70 upvotes, #1 of 2026-04-09
  18. Synthetic Sandbox for Training Machine Learning Engineering Agents 14 upvotes, #21 of 2026-04-07
  19. LagerNVS: Latent Geometry for Fully Neural Real-time Novel View Synthesis 10 upvotes, #13 of 2026-03-26
  20. Reasoning over mathematical objects: on-policy reward modeling and test time aggregation 6 upvotes, #25 of 2026-03-20
  21. Omnilingual MT: Machine Translation for 1,600 Languages 19 upvotes, #17 of 2026-03-18
  22. Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training 4 upvotes, #32 of 2026-03-13
  23. Multi-Task Reinforcement Learning for Enhanced Multimodal LLM-as-a-Judge 4 upvotes, #32 of 2026-03-13
  24. Beyond Language Modeling: An Exploration of Multimodal Pretraining 85 upvotes, #2 of 2026-03-04
  25. VecGlypher: Unified Vector Glyph Generation with Language Models 11 upvotes, #13 of 2026-02-26
  26. Decoding ML Decision: An Agentic Reasoning Framework for Large-Scale Ranking System 7 upvotes, #14 of 2026-02-24
  27. Learning Personalized Agents from Human Feedback 8 upvotes, #14 of 2026-02-19
  28. Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation 4 upvotes, #18 of 2026-02-16
  29. AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agents 70 upvotes, #6 of 2026-02-10
  30. Memorization Dynamics in Knowledge Distillation for Language Models 2 upvotes, #34 of 2026-02-02
  31. Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability 39 upvotes, #5 of 2026-01-27
  32. ActionMesh: Animated 3D Mesh Generation with Temporal 3D Diffusion 12 upvotes, #17 of 2026-01-23
  33. ShapeR: Robust Conditional 3D Shape Generation from Casual Captures 20 upvotes, #8 of 2026-01-19
  34. Inference-time Physics Alignment of Video Generative Models with Latent World Models 12 upvotes, #23 of 2026-01-16
  35. VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice 32 upvotes, #5 of 2026-01-09
  36. PhyGDPO: Physics-Aware Groupwise Direct Preference Optimization for Physically Consistent Text-to-Video Generation 18 upvotes, #9 of 2026-01-01
  37. Training AI Co-Scientists Using Rubric Rewards 17 upvotes, #13 of 2025-12-30
  38. HiStream: Efficient High-Resolution Video Generation via Redundancy-Eliminated Streaming 20 upvotes, #8 of 2025-12-25
  39. SAM Audio: Segment Anything in Audio 21 upvotes, #8 of 2025-12-24
  40. Physics of Language Models: Part 4.1, Architecture Design and the Magic of Canon Layers 22 upvotes, #9 of 2025-12-22
  41. Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image 12 upvotes, #16 of 2025-12-19
  42. Flowception: Temporally Expansive Flow Matching for Video Generation 3 upvotes, #29 of 2025-12-16
  43. OneStory: Coherent Multi-Shot Video Generation with Adaptive Memory 43 upvotes, #4 of 2025-12-10
  44. Scaling Zero-Shot Reference-to-Video Generation 28 upvotes, #6 of 2025-12-09
  45. TV2TV: A Unified Framework for Interleaved Language and Video Generation 14 upvotes, #15 of 2025-12-05
  46. TUNA: Taming Unified Visual Representations for Native Unified Multimodal Models 60 upvotes, #5 of 2025-12-02
  47. SAM 3: Segment Anything with Concepts 96 upvotes, #1 of 2025-11-24
  48. WorldGen: From Text to Traversable and Interactive 3D Worlds 18 upvotes, #8 of 2025-11-24
  49. SAM 3D: 3Dfy Anything in Images 95 upvotes, #2 of 2025-11-21
  50. Mixture of States: Routing Token-Level Dynamics for Multimodal Generation 6 upvotes, #8 of 2025-11-20
  51. What Does It Take to Be a Good AI Research Agent? Studying the Role of Ideation Diversity 54 upvotes, #3 of 2025-11-20
  52. Souper-Model: How Simple Arithmetic Unlocks State-of-the-Art LLM Performance 130 upvotes, #2 of 2025-11-18
  53. The Path Not Taken: RLVR Provably Learns Off the Principals 27 upvotes, #6 of 2025-11-12
  54. DigiData: Training and Evaluating General-Purpose Mobile Control Agents 5 upvotes, #21 of 2025-11-11
  55. CRAG-MM: Multi-modal Multi-turn Comprehensive RAG Benchmark 4 upvotes, #20 of 2025-10-31
  56. SPICE: Self-Play In Corpus Environments Improves Reasoning 12 upvotes, #25 of 2025-10-29
  57. Beyond Reasoning Gains: Mitigating General Capabilities Forgetting in Large Reasoning Models 14 upvotes, #22 of 2025-10-29
  58. The Art of Scaling Reinforcement Learning Compute for LLMs 29 upvotes, #8 of 2025-10-16
  59. HoneyBee: Data Recipes for Vision-Language Reasoners 9 upvotes, #22 of 2025-10-15
  60. The Alignment Waltz: Jointly Training Agents to Collaborate for Safety 39 upvotes, #12 of 2025-10-10
  61. Hybrid Reinforcement: When Reward Is Sparse, It's Better to Be Dense 30 upvotes, #14 of 2025-10-10
  62. OneFlow: Concurrent Mixed-Modal and Interleaved Generation with Edit Flows 12 upvotes, #15 of 2025-10-08
  63. CWM: An Open-Weights LLM for Research on Code Generation with World Models 1 upvotes, #35 of 2025-10-07
  64. TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning 47 upvotes, #6 of 2025-10-01
  65. The Era of Real-World Human Interaction: RL from User Conversations 16 upvotes, #25 of 2025-09-30
  66. DepthLM: Metric Depth From Vision Language Models 8 upvotes, #43 of 2025-09-30
  67. Jointly Reinforcing Diversity and Quality in Language Model Generations 25 upvotes, #13 of 2025-09-03
  68. DINOv3 194 upvotes, #1 of 2025-08-18
  69. VGGT: Visual Geometry Grounded Transformer 20 upvotes, #7 of 2025-03-17
  70. Cluster and Predict Latents Patches for Improved Masked Image Modeling 2 upvotes, #23 of 2025-02-17

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.