한국어EN
Evaluation & BenchmarksPercy Liang et al. (Stanford CRFM)Stanford University

Stanford CRFM releases HELM

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

Released HELM — 42 scenarios × 7 metrics × 30 models. Raised the share of evaluation scenarios shared across models from 17.9% to 96.0%

Why it matters

The work that made comparability between models possible in the first place

Source — Primary sources first; secondary reports are labelled “Press”

Paperarxiv.orghttps://arxiv.org/abs/2211.09110

Added in Week 35, 2026 · See that week

On the map — ⑯ Evaluation line

  1. Previous stopMMLU
  2. This recordHELM
  3. Next stopSWE-bench
See it on the map

Same week

Gao et al. present PAL (Program-aided Language Models)Paper

Nearby on the same axis

Srivastava et al. release BIG-bench (204 tasks)PaperLMSYS launches Chatbot ArenaOfficialMT-Bench and LLM-as-a-judge publishedPaperHugging Face explains the MMLU score mismatch on the Open LLM LeaderboardOfficial