Developers can now add live Google Maps data to Gemini-powered AI app outputs
10 月 18, 2025admin NUAI,Committee,新闻,Uncategorized(0)Google is adding a new feature for third-party developers building atop its Gemini AI models...
Detecting value-expressive text posts in Russian social media
12 月 10, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2312.08968v3 Announce Type: replace Abstract: Basic values are concepts or beliefs which pertain to desirable...
Detecting Sockpuppetry on Wikipedia Using Meta-Learning
6 月 13, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2506.10314v1 Announce Type: cross Abstract: Malicious sockpuppet detection on Wikipedia is critical to preserving access...
Detecting Hope, Hate, and Emotion in Arabic Textual Speech and Multi-modal Memes Using Large Language Models
8 月 25, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2508.15810v1 Announce Type: new Abstract: The rise of social media and online communication platforms has...
Detecting Hallucinations in Authentic LLM-Human Interactions
10 月 14, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2510.10539v1 Announce Type: new Abstract: As large language models (LLMs) are increasingly applied in sensitive...
DeSTA2.5-Audio: Toward General-Purpose Large Audio Language Model with Self-Generated Cross-Modal Alignment
3 月 20, 2026admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2507.02768v2 Announce Type: replace-cross Abstract: We introduce DeSTA2.5-Audio, a general-purpose Large Audio Language Model (LALM)...
Designing High-Performance GPU Kernels with TileLang: Tensor-Core GEMM, Fused Softmax, FlashAttention, and Autotuning
7 月 25, 2026admin NUAI,Committee,新闻,Uncategorized(0)In this tutorial, we explore TileLang as a high-level Python domain-specific language for designing and...
Designing AI Agents That Can Self-Correct
8 月 6, 2026admin NUAI,Committee,新闻,Uncategorized(0)With the vocabulary and the failure modes in place, here’s the build...
Designing AI Agents That Can Self-Correct
8 月 6, 2026admin NUAI,Committee,新闻,Uncategorized(0)With the vocabulary and the failure modes in place, here’s the build...
