512 mined sentence pairs lift Qwen's low-resource MT by +21.4 chrF++ on average
anas_ant · x · 2026-11-06
Results table from the paper thread: for 10 languages without NLLB support, Qwen's zero-shot average chrF++ of 9.36 rose to 30.76 with 512 sentence pairs mined from grammar books (+21.4 average); languages with NLLB support gained +8.90 on average. One outlier (dgo) dropped 0.50. See the main thread post for the full OCR+VLM mining pipeline.
More from Research
- Kording Lab amplifies thread: objective functions as concise descriptions of what neural nets learn — KordingLab · 2026-11-06
- LLM peer review is here to stay, researchers say — but 'excuses to reject' era must end — examachine · 2026-11-06
- UChicago names 31 researchers to 2026 Rising Stars in Data Science, with focus on world models — cindy_x_wu · 2026-11-06
- Surge researcher praises Tinker: parallel RL experiments without GPU management — echen · 2026-11-06
- Nemotron 3.5 Super + Surge RL matches GPT-5.6 Terra across 12 benchmarks — echen · 2026-11-06
- BMVA symposium on computer vision interpretability opens CFP, London, Jan 2027 — HildeKuehne · 2026-11-06