KG-to-Text 2026 - Coverage WebNLG17 GLM Judge: leaderboard
KG-to-Text study:BinEval atomic yes/no graph-grounded assessment of generated text,using DeepSeek V4.1 Flash or GLM5.3Flash judge with thinking disabled as separate protocols. Generators are the same prompted/fine-tuned cohorts as the main tables; score is a bounded mean,not judge accuracy. Mean score (0–1).
Metric: Mean score (0–1). Source: arxiv.org. Saturation forecast: Around January 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Mini (Minimal) | 1 | #189 (GPT-5 Mini) |
| 2 | Gemma 4 31B (IT) (Non-reasoning) | 1 | #179 (Gemma 4 31B (IT)) |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=kg-to-text-2026-coverage-webnlg17-glm-judge · How It Works · Data refreshed daily, snapshot 2026-10-24.