Metraipolis › archive › dc_fedfd253f5ad

MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement — arXiv:2610.11959

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.11959
License (as archived): CC0-1.0
Archived: 2026-10-09T06:13:28.094Z
SHA-256 of the archived text: 16d52cd7e21c660d27276359677957eed671318395a01dbd63c79798bf03fb17
Stable URL of this page: https://metraipolis.com/documento/dc_fedfd253f5ad

How to cite

MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement — arXiv:2610.11959 — arXiv (en), https://arxiv.org/abs/2610.11959 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fedfd253f5ad (Metraipolis archive dc_fedfd253f5ad, sha256 16d52cd7e21c660d27276359677957eed671318395a01dbd63c79798bf03fb17)

BibTeX

@misc{dcfedfd253f5ad,
  title = {MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement — arXiv:2610.11959},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.11959},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fedfd253f5ad; sha256 16d52cd7e21c660d27276359677957eed671318395a01dbd63c79798bf03fb17},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.11959. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvement. This report introduces the MiMo-V2.6 series, an omni-modal family that pushes the frontier of model intelligence by scaling RL compute. Prior to RL, we conduct mid-training on a broad multimodal corpus to provide ample exploration space, and build a solid infrastructure on the pretrained hybrid-SWA architecture to support subsequent scale-up. We scale RL compute along three dimensions: (1) larger batches and higher throughput, with an asynchronous training that consumes 1,568 samples and 2.7-3.7B tokens per step at context lengths of up to 1M; (2) more diverse and complex environments, spanning code, general, visual, and cyber domains under a mixture of agent harnesses; and (3) more grader compute, via groupwise agentic grading that yields more accurate reward signals for long-horizon tasks and steers the model towards shorter, more token-efficient solutions. To keep training stable at scale, we freeze the MoE router and establish a multi-layer defense against reward hacking. We further build infrastructure for mixed-task agentic RL, including a unified trajectory representation, high-concurrency multi-framework rollout, decoupled control and data planes, and training-inference consistency. We open-source the training dynamics, RL environments, and RL framework to facilitate reproduction and further research on scaled RL and model self-improvement.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.11959 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T05:47:15.310Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).