LLM metric

_{scroll ↓ to Resources}

Contents

• Note
  • [[#Note#For text|For text]]
    • [[#For text#For machine translation|For machine translation]]
    • [[#For text#For summarization|For summarization]]
    • [[#For text#For chatbots|For chatbots]]
    • [[#For text#Other, task-agnostic|Other, task-agnostic]]
  • [[#Note#For images|For images]]
  • [[#Note#Text-to-Video|Text-to-Video]]
• Resources

Note

For text

For machine translation

• Levenshtein Ratio
• BERTScore
• GEMBA

For summarization

• ROUGE
• G-Eval

For chatbots

• how to evaluate LLM chatbots
• moderation
• answer relevance

Other, task-agnostic

• perplexity
• hallucination

For images

• FID, IS, KID, SWD, PPL, LPIPS

Text-to-Video

• FVD, CLIPScore, FID, LPIPS, KID

Resources

• Key LLM Evaluation Metrics & How to Calculate Them

---

Links to this File

table file.inlinks, filter(file.outlinks, (x) => !contains(string(x), ".jpg") AND !contains(string(x), ".pdf") AND !contains(string(x), ".png")) as "Outlinks" from [[]] and !outgoing([[]])  AND -"Changelog"