Papers

• MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production

  Peking University

  1 author
• AI+HW 2035: Shaping the Next Decade

  Princeton University, Stanford University, State University of New York, University of California

  Published on: 2026-03-05 1 author
• GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training

  Peking University

  Published on: 2026-02-15 1 author
• GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

  Published on: 2026-02-14 1 author
• AdaptEvolve: Improving Efficiency of Evolutionary AI Agents through Adaptive Model Selection

  Indian Institute of Technology Kharagpur

  Published on: 2026-02-12 1 author
• Canzona: A Unified, Asynchronous, and Load-Balanced Framework for Distributed Matrix-based Optimizers

  Published on: 2026-02-04 1 author
• M2XFP: A Metadata-Augmented Microscaling Data Format for Efficient Low-bit Quantization

  Shanghai Jiao Tong University

  Published on: 2026-01-28 1 author
• Zebra-Llama: Towards Extremely Efficient Hybrid Models

  Published on: 2026-01-20 1 author
• Power Aware Dynamic Reallocation For Inference

  Published on: 2026-01-18 1 author
• AIE4ML: An End-to-End Framework for Compiling Neural Networks for the Next Generation of AMD AI Engines

  Institute of Physics Belgrade

  Published on: 2026-01-09 1 author
• CD4LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models

  Princeton University

  Published on: 2026-01-05 1 author
• CXL-SpecKV: A Disaggregated FPGA Speculative KV-Cache for Datacenter LLM Serving

  Columbia University, Yale University

  Published on: 2025-12-11 2 authors
• SparK: Query-Aware Unstructured Sparsity with Recoverable KV Cache Channel Pruning

  University of Chinese Academy of Sciences

  Published on: 2025-10-12 1 author
• Efficient and Adaptable Overlapping for Computation and Communication via Signaling and Reordering

  Tsinghua University

  Published on: 2025-10-09 1 author
• APRIL: Active Partial Rollouts in Reinforcement Learning to Tame Long-tail Generation

  Carnegie Mellon University

  Published on: 2025-09-26 1 author
• Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks

  Published on: 2025-07-31 1 author
• Agent Laboratory: Using LLM Agents as Research Assistants

  ETH Zurich

  Published on: 2025-06-17 1 author