Research
PATMT2: Two-stage domain adaptation for multilingual patent translation
arXiv:2610.16466v1 Announce Type: cross Abstract: We present PatMT2 (Patent Machine Translation, v2), a two-stage LoRA pipeline that adapts the Hy-MT2-7B dedicated translation model to four-language (
arXiv:2610.16466v1 Announce Type: cross Abstract: We present PatMT2 (Patent Machine Translation, v2), a two-stage LoRA pipeline that adapts the Hy-MT2-7B dedicated translation model to four-language (Chinese/English/Japanese/Korean) patent translation. On six translation directions, the Stage-2 checkpoint reaches a macro-average BLEU of 43.39 (BF16) to 43.57 (AWQ INT4), improving over the Hy-MT2-7B base by up to +8.8 BLEU and also exceeding Qwen3-32B and several external systems on the same average. We then measure the cost of adaptation: probing 38 languages absent from the fine-tuning data, the model still produces correct, fluent output on 34, while four low-resource languages collapse into distinct, reproducible failure modes. We treat these as diagnostic cases rather than retention rates, and report their mechanisms without attributing them to a single cause. Paired bootstrap tests confirm the translation gains over all baselines (p < 0.001).
Related
- Forward Pass Domain Adaptation (Without Cross-Layer Backpropagation)
- DomainPilot: Domain-Level Loss-Guided Two-Stage Data Mixture Optimization for Efficient Language Model Fine-Tuning
- AR1-ZO: Topology-Aware Rank-1 Zeroth-Order Queries for High-Rank LoRA Fine-Tuning
- Breaking the Structural Identity: Personalized Federated LoRA Fine-tuning under Rank Heterogeneity
Source: arXiv cs.AI | 2026-10-15