Dev suspects longer training degraded Qwen backbone multilingual abilities, plans more multilingual data
antoine_chaffin · x · 2026-10-31
Antoine Chaffin responds to a challenge about Wolof language performance, saying it's almost certainly a base-backbone issue rather than lack of targeted training: both models likely use a Qwen backbone. If the gap is large, longer training may have cost some multilingual capability, and he plans to add multilingual data — though he doubts it will lift Wolof much. He also plans to try translation-enabled generalization (mLateOn style) and is asking for public Wolof evals to probe whether the base backbone degraded.
More from Models
- Anthropic removes Claude Ultra's 5-hour limit, keeps weekly cap — umeshonai · 2026-10-31
- 5 cheap open models vote vs Opus: 89.6% accuracy at 1/10th the cost — PomegranateAble4339 · 2026-10-31
- Gemini 4 Carbon reportedly scores 53 FCI on FrontierCS 2, big jump from Argon in weeks — clmt · 2026-10-31
- Gemini 3.5 Pro reportedly delayed into oblivion while rivals took share, says dev — tarantulae · 2026-10-31
- LiquidAI's LFM3-VL d1-2B edge vision-language model trends on Hugging Face — LiquidAI · 2026-10-31
- OpenAI's new API credits make ChatGPT Pro tiers a no-brainer for API users — Quiet-Recording-9269 · 2026-10-31