Preprint compares Banach, Hilbert and Euclidean domain adaptation
An L1 density regularizer beats an identical L2 ablation on every reported task, including 92.44% versus 89.80% on USPS to MNIST.
Stories per day · 30 days
1 RSS
Relatedbanach1density-estimation1hanja1historical-archives1machine-translation1manifold-learning1parametric-memory1pseudo-labeling1question-answering1rag1retrieval1transfer-learning1
9 stories, newest first · one card per story
Stories tagged #domain-adaptation
HERIT adapts translation models for Hanja-to-English historical archives
Authors use RAG on Hanja-Korean corpora to build pseudo-labels, cut temporal bias, and report gains over baselines.
Deep Fusion Memory adds chunk-level parametric memory to LLMs
With the backbone frozen, training needs 32 times fewer retrieval queries and inference uses no explicit retrieval.
HuatuoGPT-3: RL-Only Domain Adaptation from Base Models
OnePO does RL-only domain adaptation and produces HuatuoGPT-3, whose 27B model leads reported HealthBench scores.
Chinese-Jev: Bringing System One Model to Chinese-Language Tasks
Chinese-Jev, a lightweight encoder-only decision model, beats closed-source Jev on Chinese tasks with up to 20x speedup.
Preference-Guided Adaptation for Open-Vocabulary Semantic Segmentation via Prompt Disagreement
A preference-guided method adapts open-vocabulary segmentation using prompt disagreement, improving MESS results without pixel masks.
Training Specialist Models without Reasoning Trajectories for Domain Expert Distillation
Study shows specialists trained on question-answer pairs implicitly select latent reasoning trajectories, and tuning choices control the precision-generalization trade-off in distillation.
Transfer Learning for Evolving Domains
TrED formalizes transfer learning for domains whose data availability evolves over time, arguing classical settings are regimes along one trajectory, and remains unsolved.
Climate-ModernBERT: Revisiting Corpus Composition for Domain-Adaptive Continued Pretraining
Climate-ModernBERT domain-adapted encoders reach 76.3 average F1 across nine climate benchmarks, 2.8 points above vanilla ModernBERT-Base.