cd/entity/github.com/megagonlabs/mawile-judgeĀ· homeā€ŗ entitiesā€ŗ github.com/megagonlabs/mawile-judge
grep -l @github.com/megagonlabs/mawile-judge /news/*.json | wc -l ā†’ 1

github.com/megagonlabs/mawile-judge

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

04:00
2026-09-23
arxiv.org
large-language-models

MAWILE: Multi-Axis Workbench for Inspecting LLM Evaluators

Megagon Labs researchers released MAWILE, a developer-facing workbench for auditing the sensitivity of large language model judges across four surfaces: the judge prompt, judge rubric, target-system iā€¦

// co-occurs with top 3 entities
Megagon Labs 1 MAWILE 1 arXiv 1
// topics top 5 topics
large language models 1 ai research 1 ai tools 1 developer tools 1 ai safety 1