Back to AIBriefs
AnalysisAI Models

Together AI unveils 9 research papers at ICML 2026

Together AI
@togethercompute

Accelerate inference, model shaping, and pre-training on a research-optimized platform.

San Francisco, CAtogether.ai
View on X
Together AI avatar
Together AI
@togethercompute

7/ Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking Paper: https://t.co/LgGqu8vl97

Together AI unveils 9 research papers at ICML 2026
6/ When RL Meets Adaptive Speculative Training: A Unified Training-Serving System (Aurora) Paper:...2 hours agoTogether AI
9/ ParallelKernelBench: Benchmarking LLMs on Multi-GPU Kernel Generation Paper:...2 hours agoTogether AI
4/ Escaping the Verifier: Learning to Reason via Demonstrations (RARO) Paper:...2 hours agoTogether AI
Our research team has 9 papers at ICML next week! Spanning the full stack from frontier agents to...2 hours agoTogether AI
Read the blog: See us at ICML:2 hours agoTogether AI
1/ DSGym: A Holistic Framework for Evaluating and Training Data Science Agents Paper:...2 hours agoTogether AI
2/ ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System Paper:...2 hours agoTogether AI
8/ Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining...2 hours agoTogether AI
3/ Learning to Discover at Test Time (TTT-Discover) Paper:2 hours agoTogether AI
5/ V1: Unifying Generation and Self-Verification for Parallel Reasoners Paper:...2 hours agoTogether AI
·
3 hours ago
Together AI unveils 9 research papers at ICML 2026 — AIBriefs