arXiv Proposes CAFE: Revolutionizing Feature Distillation in Multilingual Speech Recognition
arXiv's latest research introduces CAFE (Confined-Adapter Feature distillation), a novel a…
ZICQ curates daily AI news and deep dives on models, open-source infrastructure, and developer field reports.
Submit an articleCurrent filter: Tag: #历史文本处理 Clear filters
arXiv's latest research introduces CAFE (Confined-Adapter Feature distillation), a novel a…
LLMs & Foundation Models· 2026-10-15
arXiv has released DecisionRobust, a novel benchmark for systematically evaluating the rob…
Agentic· 2026-10-15
A new study on arXiv proposes a hybrid architecture where cloud-written programs defer bou…
LLMs & Foundation Models· 2026-10-15
arXiv introduces MOSAIC, a novel method for cross-modal training sample attribution aimed …
LLMs & Foundation Models· 2026-10-15
arXiv introduces SQUELCH, a novel approach to enhance low-rank adaptation (LoRA) by dynami…
Research & Papers· 2026-10-15
arXiv has released a new study on game theory algorithms, introducing IR-PRM+, an enhanced…
Research & Papers· 2026-10-15
A new study published on arXiv introduces a method for improving indoor radio-map predicti…
LLMs & Foundation Models· 2026-10-15
arXiv presents a study on data density scaling laws for image self-distillation models, de…
Research & Papers· 2026-10-15
arXiv introduces PMO-Dock, a benchmark and protocol for molecular optimization focused on …
LLMs & Foundation Models· 2026-10-15
A new study on arXiv introduces Fisher Utility Weighting (FUW), a novel token selection me…
Research & Papers· 2026-10-15
arXiv has released a study on machine learning models for predicting the effects of gene c…
Research & Papers· 2026-10-15
A new arXiv study investigates the mechanisms of in-context learning (ICL) in language mod…