Rosetta at AlexandriaX-2026: LoRA-Adapted NileChat for Context-Aware Dialectal Arabic Dialogue Translation
Rosetta ranks 4th and 5th in AlexandriaX-2026 dialectal Arabic dialogue translation using a LoRA adapter on NileChat-3B, finding limited pretraining benefit.
The Rosetta system for the AlexandriaX-2026 shared task fine-tunes a LoRA adapter on NileChat-3B for context-aware English-to-dialectal Arabic dialogue translation. The adapter was additionally pretrained on MADAR and PADIC dialect corpora for the unconstrained track. It achieved spBLEU 26.10 (4th, constrained) and 25.09 (5th, unconstrained). External dialect pretraining improved only two of thirteen dialects while slightly degrading overall performance, indicating negative transfer.
- LoRA fine-tuning of NileChat-3B conditioned on dialect and dialogue context via structured prompts
- Additional MADAR/PADIC pretraining helped only 2 of 13 dialects
- Ranks 4th (constrained) and 5th (unconstrained) with spBLEU ~26/25
- Evidence of negative transfer from external dialect pretraining
- Organizations
- Rosetta
- AI models
- NileChat-3B
From the original article
1 min read at arxiv.orgThis paper describes the Rosetta system for Subtask 1 (Context-Aware English-to-Dialectal Arabic Dialogue Translation) of the AlexandriaX shared task, participating in both constrained and unconstrained tracks. The approach fine-tunes a LoRA adapter on NileChat-3B using structured system/user prompts that condition generation on dialect and dialogue context.
By Nada Esmaeil · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.