Papers

Filter by company

FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling

Meta Platforms, NVIDIA, Google, Together AI / Princeton University

Published on: 2026-03-05 1 author
V1 : Unifying Generation and Self-Verification for Parallel Reasoners

NVIDIA, Together AI / UC Berkeley

Published on: 2026-03-04 1 author
Speculative Speculative Decoding

Together AI / Stanford University

Published on: 2026-03-03 1 author
Learning to Discover at Test Time

Together AI, NVIDIA / Stanford University, UC San Diego

Published on: 2026-02-05 1 author
Asynchronous Reasoning: Training-Free Interactive Thinking LLMs

Together AI / The University of Tokyo

Published on: 2026-02-04 1 author
DSGym: A Holistic Framework for Evaluating and Training Data Science Agents

Together AI / Stanford University

Published on: 2026-01-22 1 author
Understanding and Steering the Cognitive Behaviors of Reasoning Models at Test-Time

Together AI / The University of Texas

Published on: 2026-01-19 1 author
MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding

Together AI / Carnegie Mellon University

Published on: 2025-04-02 1 author
RedPajama: an Open Dataset for Training Large Language Models

Together AI, EleutherAI / Stanford University, The Ohio State University

Published on: 2024-10-19 1 author

Search