Accent-Invariant Automatic Speech Recognition via Saliency-Driven Spectrogram Masking
10 月 13, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2510.09528v1 Announce Type: new Abstract: Pre-trained transformer-based models have significantly advanced automatic speech recognition (ASR)...
Accelerating Transformer Training with NVIDIA Transformer Engine, Fused Kernels, BF16, FP8, and GPU Benchmarking
8 月 1, 2026admin NUAI,Committee,新闻,Uncategorized(0)In this tutorial, we explore how NVIDIA Transformer Engine accelerates transformer workloads by combining fused...
Accelerating Diffusion LLMs via Adaptive Parallel Decoding
11 月 3, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2506.00413v2 Announce Type: replace Abstract: The generation speed of LLMs are bottlenecked by autoregressive decoding...
AbstRaL: Teaching LLMs Abstract Reasoning via Reinforcement to Boost Robustness on GSM Benchmarks
7 月 6, 2025admin NUAI,Committee,新闻,Uncategorized(0)Recent research indicates that LLMs, particularly smaller ones, frequently struggle with robust reasoning. They tend...
Abstract or die: Why AI enterprises can’t afford rigid vector stacks
10 月 19, 2025admin NUAI,Committee,新闻,Uncategorized(0)Vector databases (DBs), once specialist research instruments, have become widely used infrastructure in just a...
A woman’s uterus has been kept alive outside the body for the first time
3 月 28, 2026admin NUAI,Committee,新闻,Uncategorized(0)“Think of this as a human body,” says Javier González. In front of me is...
A Unified Representation for Continuity and Discontinuity: Syntactic and Computational Motivations
6 月 9, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2506.05686v1 Announce Type: new Abstract: This paper advances a unified representation of linguistic structure for...
A Tale of Two Scripts: Transliteration and Post-Correction for Judeo-Arabic
1 月 31, 2026admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2507.04746v2 Announce Type: replace Abstract: Judeo-Arabic refers to Arabic variants historically spoken by Jewish communities...
A Systematic Study of Pseudo-Relevance Feedback with LLMs
3 月 12, 2026admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2603.11008v1 Announce Type: cross Abstract: Pseudo-relevance feedback (PRF) methods built on large language models (LLMs)...
