Category

Translation: RL environments and AI benchmarks

AI benchmarks
22
Shares entries with
0 categories

Snapshot 2026-10-11. How categories are built · Where the data comes from

In short

What are RL environments for translation?
A simulated workplace where an AI agent practises a task (a code repository, a browser, business software) and an automatic grader scores each attempt. Labs train on thousands of these attempts. Bought to improve models.

Only benchmarks use the Translation tag: the RL environment directories file their vendors under other categories. Browse all RL environments.

AI benchmarks 22

None confirmed yet.

22 not yet confirmedListed in the catalog; no line from its own site or paper confirms it yet.
  1. VNTL Leaderboard
  2. JP-TL-Bench
  3. CULTURE-MT
  4. HardMTBench
  5. Beyond Literal Mapping
  6. Reeve Foundation Multilingual Corpus
  7. WMT24++
  8. AfriScience-MT
  9. PaliBench
  10. STEB
  11. Alexandria
  12. Cultural Fidelity in English-to-Hindi Translation
  13. Sand-Glass
  14. Translation en→Set1 COMET22
  15. Translation en→Set1 spBleu
  16. Translation Set1→en COMET22
  17. Translation Set1→en spBleu
  18. Nsanku
  19. On the Cultural Anachronism and Temporal Reasoning in Vision Language Models
  20. An Empirical Study of On-Device Translation for Real-Time Live-Stream Chat on Mobile Devices
  21. VIDA
  22. WMT23

Next: pick another domain. Tags come from the RL Research daily scrape, canonicalized nightly.