hAIgopala › AI terms › Rapid Fire (3 terms per short) › Day 36

BLEU / ROUGE

Evaluation & MetricsChapter 10: Report Card for BitTerm 177 of 253

Metrics comparing generated text with reference text (translation / summarisation).

At Sunrise School

Gopal asks

How do we score translations and summaries?

Bit answers

BLEU and ROUGE compare word overlap with a reference written by staff. Quick, but they miss meaning, so we add a judge too.

In the real world

Word-overlap scores.

Watch the short

Short coming soonSeason 5 · Day 36Follow @haigopala

Quick check

Quick check. Do public benchmarks prove which model is best for our school? Yes or no?

Taught together with

Benchmark · A/B Testing

Practise it

Try this term and all 253 others in the free Model Training Arena.

Open the Arena →

Follow Bit daily: YouTube English · తెలుగు · हिंदी  |  Instagram English · తెలుగు · हिंदी