cd /news/machine-learning/hugging-face-fixes-trl-cispo-default… · home › topics › machine-learning › article
[ARTICLE · art-169535] src=runtimewire.com ↗ pub= topic=machine-learning verified=true sentiment=· neutral

Hugging Face fixes TRL CISPO default that shrank gradients fivefold in its reproduction

Hugging Face shipped TRL v1.18.0 on November 13th, 2026, raising the CISPO default for the 'epsilon_high' setting to 5.0 after an inherited 0.2 ceiling clipped every on-policy importance weight to 0.2 when users left the parameter unset, shrinking gradients fivefold in its reproduction. The fix corrects a reinforcement-learning loss default that had been silently capping on-policy weights near 1.

read1 min views2 publishedNov 14, 2026
Hugging Face fixes TRL CISPO default that shrank gradients fivefold in its reproduction
Image: Runtimewire (auto-discovered)

Hugging Face's TRL v1.18.0 sets the CISPO default to 5.0 after an inherited 0.2 ceiling clipped on-policy weights near 1.

        By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)
        · Published 

Hugging Face, co-founded by Clement Delangue, has corrected a default in TRL's CISPO reinforcement-learning loss that clipped every on-policy importance weight to 0.2 when users left 'epsilon_high' unset. The fix shipped in TRL v1.18.0 on November 13th, 2026, and raises the CISPO default for that setting to 5.0.…

── more in #machine-learning 4 stories · sorted by recency
── more on @hugging face 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/hugging-face-fixes-t…] indexed:0 read:1min 2026-11-14 · —