Re 1/ DSGym: A Holistic Framework for Evaluating and Training Data Science Agents Paper: https://arxiv.org/abs/2601.16344
Re 1/ DSGym: A Holistic Framework for Evaluating and Training Data Science AgentsPaper: https://arxiv.org/abs/2601.16344
Latest
Re Read the blog: https://www.together.ai/blog/icml-2026 See us at ICML: https://www.together.ai/icml-2026
Re Read the blog: https://www.together.ai/blog/icml-2026See us at ICML: https://www.together.ai/icml-2026
Our research team has 9 papers at ICML next week! Spanning the full stack from frontier agents to GPU kernels, we're excited to share what our researc…
Our research team has 9 papers at ICML next week! Spanning the full stack from frontier agents to GPU kernels, we're excited to share what our…
RT Zain: We gave a 2 hr deepdive on how to build inference engines that handle trillion token agentic workloads at @aiDotEngineer. Will drop slides an…
RT ZainWe gave a 2 hr deepdive on how to build inference engines that handle trillion token agentic workloads at @aiDotEngineer.Will drop slides and detailed walkthrough!
RT Victor Su-Ortiz: Missed our talk with @togethercompute on @MiniMax_AI sparse attention and kernel optimizations? Catch us tomorrow at 10:45am PT.
RT Victor Su-OrtizMissed our talk with @togethercompute on @MiniMax_AI sparse attention and kernel optimizations?Catch us tomorrow at 10:45am PT.
Re 8/ Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining (OEA) https://arxiv.org/abs/2511.02237
Re 8/ Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining (OEA)https://arxiv.org/abs/2511.02237
Re 5/ V1: Unifying Generation and Self-Verification for Parallel Reasoners https://arxiv.org/abs/2603.04304
Re 5/ V1: Unifying Generation and Self-Verification for Parallel Reasonershttps://arxiv.org/abs/2603.04304
Re 6/ When RL Meets Adaptive Speculative Training: A Unified Training-Serving System (Aurora) https://arxiv.org/abs/2602.06932
Re 6/ When RL Meets Adaptive Speculative Training: A Unified Training-Serving System (Aurora)https://arxiv.org/abs/2602.06932
Re 7/ Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking https://arxiv.org/abs/2602.21196
Re 7/ Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunkinghttps://arxiv.org/abs/2602.21196
Re 4/ Escaping the Verifier: Learning to Reason via Demonstrations (RARO) https://arxiv.org/abs/2511.21667
Re 4/ Escaping the Verifier: Learning to Reason via Demonstrations (RARO)https://arxiv.org/abs/2511.21667