SECAI Core · phase 14 of 15 · noun

Recall-oriented understudy for gisting evaluations

A set of metrics for evaluating how good a system-generated summary or text is by comparing it to one or more human (reference) summaries, with a focus on how much of the reference content is included

The Explain card

Plain English
ROUGE is a family of metrics for scoring machine-generated summaries by how much of the content in human-written reference summaries they capture, with the emphasis on recall.
Example
A summarisation model is evaluated on incident reports. Its ROUGE scores stay high, but reviewers find it fabricating details, showing that overlap with references does not guarantee accuracy.
Why it matters
Defenders increasingly rely on AI summaries of logs and alerts. ROUGE helps measure whether a model keeps the important facts, but it does not catch hallucination, so pair it with human review.
Hook
Did the summary remember the things that mattered?

Word knowledge

How the word is built, where it came from, and what it sits beside in memory.

In a sentence

The red-team report places recall-oriented understudy for gisting evaluations scores beside factuality checks and prompt-injection resistance results for each summarization model.

Why these words
  • recall-oriented English recall, from re- plus call, joined to oriented identifies recall as the metric family's guiding emphasis
  • understudy English theatrical compound of under plus study presents automated scoring as a stand-in for human judgment
  • for Old English for links the stand-in role to its intended task
  • gisting English gist plus -ing; gist comes from Anglo-French gist, meaning it lies names the activity of capturing a text's gist
  • evaluations French évaluer and English -ation, with plural -s marks instances or forms of assessment
Where it came from
Origin
English acronymic coinage by Chin-Yew Lin and Eduard Hovy, formed to yield ROUGE from Recall-Oriented Understudy for Gisting Evaluation
Entered the language
2003
What changed
The theatrical sense of understudy was extended metaphorically to automatic scoring, and the original singular Evaluation later acquired the ordinary plural evaluations in some uses.
How it is spelled
Pattern
Recall-oriented follows the regular hyphenated pattern used when an -oriented compound modifies a following noun.
Pattern
Gisting adds -ing without doubling the final t because gist ends in the consonant cluster st.
Pattern
Evaluations forms its plural with regular final -s.
Breaks the pattern
The canonical expansion of ROUGE uses singular Evaluation, although the supplied phrase uses plural evaluations.

Spelled like

  • task-oriented
  • goal-oriented
  • listing
  • evaluations
Broken into chunks
  • re-call
    • revisit
    • rewrite
  • orient-ed
    • orientation
    • disoriented
  • under-study
    • underpass
    • undergraduate
  • gist-ing
    • gist
    • gists
  • evalu-ation-s
    • evaluate
    • evaluative
What it sits beside

Same subject

  • BLEU
  • METEOR
  • BERTScore
  • ROUGE-N
  • ROUGE-L

Same shape

  • task-oriented assessment
  • goal-oriented evaluation
  • human-centered evaluation

automatic text evaluation metrics

  • BLEU
  • METEOR
  • BERTScore

ROUGE variants

  • ROUGE-N
  • ROUGE-L
  • ROUGE-S

Where it sits in the deck

Phase 14: Incident Response, Evaluation, and Knowledge Bases

When detection fires, teams need structured response plans, scoring frameworks, and curated knowledge bases to triage, measure, and learn from AI security events.

Neighbours in this phase