LongStraw: Million-Token Long-Context RL

mindlab-research · hf · 2026-07-17

LongStraw proposes an execution stack tailored for million-token level RL post-training, aiming to perform long-context reinforcement learning within a fixed GPU budget.

The core idea is to perform an autograd-free evaluation of the shared prompt first, retaining only the model states required for subsequent tokens. Short response branches are then replayed sequentially, compressing the training graph and peak VRAM usage at the cost of increased replay time. The authors implemented this using Qwen3.6-27B (featuring hybrid recurrent and full attention) and GLM-5.2 (with compressed attention MoE).

Experimental results include:

The authors emphasize that these experiments primarily demonstrate execution capability rather than full training correctness, as the captured prompt state is detached, and some distributed forward/gradient combination paths are not yet complete.

Original post →

More from Infra

Infra channel →