MultiLexNorm++ (LLM Pipeline): leaderboard
Metric: Error reduction rate (%; (correct minus wrong normalizations) / words that need normalization, so 0 equals leaving the text unchanged and a negative score means more wrong than right edits; the floor is -100 divided by the share of words needing normalization; 8-shot prompting, mean of three runs; the LLM normalizes the words an XLM-R detector flags after a training-lexicon lookup; unweighted mean over 17 language settings). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 53.53 |
| 2 | Gemma 3 27B (IT) | 50.57 |
| 3 | Qwen 2.5 72B Instruct | 47.85 |
| 4 | Llama 3.3 70B Instruct | 46.22 |
Interactive version: theaggregate.ai/benchmark?slug=multilexnorm-plus-plus-llm-pipeline · How It Works · Data refreshed daily, snapshot 2026-09-26.