benchmark · first seen 17 Oct 2026
BLEU
Also: CIDEr, bilingual evaluation understudy, automatic caption evaluation metrics
An n-gram overlap metric for comparing generated text against reference texts.
Mentions per week
PapersCodeDiscussion
Earliest sources
- How Well Can LLMs Verbalize Mathematical Formulas? A Multi-Domain EvaluationPapers · 17 Oct 2026 · Peter Volbach, Tim Schlippe
- Similar Scores, Opposite Failures: Human Correction as an Instrument for Video Caption EvaluationPapers · 28 Oct 2026 · Ankit Khedia, Astitva Jaiswal, Chaitanya Prasad, Anisha Raju, Aangeeras
Latest sources
- Similar Scores, Opposite Failures: Human Correction as an Instrument for Video Caption EvaluationPapers · 28 Oct 2026 · Ankit Khedia, Astitva Jaiswal, Chaitanya Prasad, Anisha Raju, Aangeeras
- How Well Can LLMs Verbalize Mathematical Formulas? A Multi-Domain EvaluationPapers · 17 Oct 2026 · Peter Volbach, Tim Schlippe