Echoes of BERT: Do Modern Language Models Rediscover the Classical NLP Pipeline?
10月 17, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2506.02132v4 Announce Type: replace Abstract: Large transformer-based language models dominate modern NLP, yet our understanding...
Echoes of Automation: The Increasing Use of LLMs in Newsmaking
8月 15, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2508.06445v2 Announce Type: replace Abstract: The rapid rise of Generative AI (GenAI), particularly LLMs, poses...
EBPO: Empirical Bayes Shrinkage for Stabilizing Group-Relative Policy Optimization
2月 6, 2026admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2602.05165v1 Announce Type: cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective for...
DySK-Attn: A Framework for Efficient, Real-Time Knowledge Updating in Large Language Models via Dynamic Sparse Knowledge Attention
12月 29, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2508.07185v2 Announce Type: replace Abstract: Large Language Models (LLMs) suffer from a critical limitation: their...
Dynamic Context-Aware Streaming Pretrained Language Model For Inverse Text Normalization
6月 2, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2505.24229v1 Announce Type: new Abstract: Inverse Text Normalization (ITN) is crucial for converting spoken Automatic...
Dynamic Acoustic Model Architecture Optimization in Training for ASR
6月 19, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2506.13180v2 Announce Type: replace Abstract: Architecture design is inherently complex. Existing approaches rely on either...
DVAGen: Dynamic Vocabulary Augmented Generation
10月 21, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2510.17115v1 Announce Type: new Abstract: Language models trained with a fixed vocabulary struggle to generalize...
DualDistill and Agentic-R1: How AI Combines Natural Language and Tool Use for Superior Math Problem Solving
7月 25, 2025admin NUAI,Committee,ニュース,Uncategorized(0)Existing long-CoT reasoning models have achieved state-of-the-art performance in mathematical reasoning by generating reasoning trajectories...
DraCo: Draft as CoT for Text-to-Image Preview and Rare Concept Generation
12月 5, 2025admin NUAI,Committee,ニュース,Uncategorized(0)arXiv:2512.05112v1 Announce Type: cross Abstract: Recent unified multimodal large language models (MLLMs) have shown impressive...
