Papers

• Lost in the Middle at Birth: An Exact Theory of Transformer Position Bias

  Published on: 2026-03-10 1 author
• ReMix: Reinforcement routing for mixtures of LoRAs in LLM finetuning

  University of Illinois at Urbana-Champaign, Washington University

  Published on: 2026-03-10 18 authors
• Robust Post-Training for Generative Recommenders: Why Exponential Reward-Weighted SFT Outperforms RLHF

  Stanford University

  Published on: 2026-03-10 5 authors
• Towards a Neural Debugger for Python

  Johannes Kepler University Linz

  Published on: 2026-03-10 4 authors
• DARC: Disagreement-Aware Alignment via Risk-Constrained Decoding

  Fudan University, Harbin Institute of Technology

  Published on: 2026-03-09 7 authors
• Offline Materials Optimization with CliqueFlowmer

  University of California

  Published on: 2026-03-06 4 authors
• CHMv2: Improvements in Global Canopy Height Mapping using DINOv3

  University of Maryland

  Published on: 2026-03-06 12 authors
• FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling

  Colfax Research, Georgia Tech, Princeton University

  Published on: 2026-03-05 6 authors
• FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling

  Princeton University

  Published on: 2026-03-05 1 author
• T2S-Bench & Structure-of-Thought: Benchmarking and Prompting Comprehensive Text-to-Structure Reasoning

  Duke University

  Published on: 2026-03-04 1 author
• InfinityStory: Unlimited Video Generation with World Consistency and Character-Aware Shot Transitions

  University of Memphis, University of Oregon

  Published on: 2026-03-04 1 author
• Beyond Language Modeling: An Exploration of Multimodal Pretraining

  New York University

  Published on: 2026-03-03 1 author
• Agentic Code Reasoning

  Published on: 2026-03-02 1 author
• Compositional Planning with Jumpy World Models

  McGill University

  Published on: 2026-02-23 1 author
• Wink: Recovering from Misbehaviors in Coding Agents

  Published on: 2026-02-20 1 author
• SARAH: Spatially Aware Real-time Agentic Humans

  Published on: 2026-02-20 1 author
• Image Generation with a Sphere Encoder

  University of Maryland

  Published on: 2026-02-16 1 author
• Learning to Reason in 13 Parameters

  Cornell University

  Published on: 2026-02-04 1 author
• An Empirical Study on Noisy Data and LLM Pretraining Loss Divergence

  University of Oxford

  Published on: 2026-02-02 1 author
• ReasonCACHE: Teaching LLMs To Reason Without Weight Updates

  Massachusetts Institute of Technology

  Published on: 2026-02-02 1 author
• Agentic Very Long Video Understanding

  University of Wisconsin-Madison

  Published on: 2026-01-26 1 author
• Unified Text-Image Generation with Weakness-Targeted Post-Training

  University of Texas

  Published on: 2026-01-21 1 author
• Learning Latent Action World Models In The Wild

  National Institute for Research in Digital Science and Technology, New York University

  Published on: 2026-01-20
• Agentic Reasoning for Large Language Models

  Amazon Research, Google DeepMind, Meta, University of Illinois Urbana-Champaign

  Published on: 2026-01-18 1 author
• KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta

  Published on: 2026-01-16 1 author
• VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice

  King Abdullah University of Science and Technology (KAUST), Princeton University

  Published on: 2026-01-08 1 author
• Diffusion Forcing for Multi-Agent Interaction Sequence Modeling

  University of California

  Published on: 2025-12-19 1 author
• Physics of Language Models: Part 4.1, Architecture Design and the Magic of Canon Layers

  Published on: 2025-12-19 1 author
• GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluatio

  Allen Institute for AI, University of California, University of Washington

  Published on: 2025-12-18 1 author
• World Models Can Leverage Human Videos for Dexterous Manipulation

  New York University

  Published on: 2025-12-15 1 author
• Confucius Code Agent: Scalable Agent Scaffolding for Real-World Codebases

  Harvard University

  Published on: 2025-12-11 11 authors
• TUNA: Taming Unified Visual Representations for Native Unified Multimodal Models

  King Abdullah University of Science and Technology (KAUST), The University of Hong Kong, University of Waterloo

  Published on: 2025-12-01 1 author
• UMA: A Family of Universal Models for Atoms

  Carnegie Mellon University

  Published on: 2025-06-30 1 author
• Transformers without Normalization

  New York University

  Published on: 2025-06-14 5 authors
• Multi-Token Attention

  Published on: 2025-06-11 4 authors
• VGGT: Visual Geometry Grounded Transformer

  University of Oxford

  Published on: 2025-05-14 6 authors
• Perception Encoder: The best visual embeddings are not at the output of the network

  Ritsumeikan Global Innovation Research Organization

  Published on: 2025-04-28 1 author
• Scaling Language-Free Visual Representation Learning

  New York University, Princeton University

  Published on: 2025-04-01 11 authors
• MLGym: A New Framework and Benchmark for Advancing AI Research Agents

  University of California

  Published on: 2025-02-20 1 author
• The Llama 3 Herd of Model

  Published on: 2024-10-23 1 author
• MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases

  Published on: 2024-02-22 12 authors
• DINOv2: Learning Robust Visual Features without Supervision

  Meta AI Research

  Published on: 2024-02-02 1 author
• Llama 2: Open Foundation and Fine-Tuned Chat Models

  Published on: 2023-07-19 1 author
• IMAGEBIND: One Embedding Space To Bind Them Al

  Meta AI Research

  Published on: 2023-05-31 1 author
• Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture

  Meta AI Research

  Published on: 2023-05-13 1 author
• Segment Anything

  Meta AI Research

  Published on: 2023-05-05 1 author
• LLaMA: Open and Efficient Foundation Language Models

  Published on: 2023-02-27 10 authors
• Toolformer: Language Models Can Teach Themselves to Use Tools

  Meta AI Research

  Published on: 2023-02-09 1 author
• Flow Matching for Generative Modeling

  Weizmann Institute of Science

  Published on: 2023-02-08 1 author
• wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations

  Facebook AI Research

  Published on: 2022-10-22 1 author