Static vs. Dynamic vs. Continuous Batching in LLM Inference
8 月 4, 2026admin NUAI,Committee,新闻,Uncategorized(0)In this article, you will learn how static, dynamic, and continuous batching work in LLM...
Stateful vs. Stateless Agent Design: Tradeoffs for Scalable Agentic Systems
7 月 24, 2026admin NUAI,Committee,新闻,Uncategorized(0)In this article, you will learn how an agent’s approach to managing state — stateless...
Start Making Sense(s): A Developmental Probe of Attention Specialization Using Lexical Ambiguity
12 月 1, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2511.21974v1 Announce Type: new Abstract: Despite an in-principle understanding of self-attention matrix operations in Transformer...
STAR-Bench: Probing Deep Spatio-Temporal Reasoning as Audio 4D Intelligence
12 月 1, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2510.24693v2 Announce Type: replace-cross Abstract: Despite rapid progress in Multi-modal Large Language Models and Large...
Stanford Researchers Introduce TRACE: A Capability-Targeted Agentic Training System That Turns Recurrent Agent Failures Into Synthetic RL Environment
7 月 13, 2026admin NUAI,Committee,新闻,Uncategorized(0)Agentic LLMs often fail the same way, again and again. A Stanford research team traced...
Stand Up for Research, Innovation, and Education
10 月 25, 2025admin NUAI,Committee,新闻,Uncategorized(0)Right now, MIT alumni and friends are voicing their support for: America’s scientific and technological...
Stacked from One: Multi-Scale Self-Injection for Context Window Extension
3 月 6, 2026admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2603.04759v1 Announce Type: new Abstract: The limited context window of contemporary large language models (LLMs)...
Stable Signer: Hierarchical Sign Language Generative Model
12 月 4, 2025admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2512.04048v1 Announce Type: cross Abstract: Sign Language Production (SLP) is the process of converting the...
Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR
5 月 18, 2026admin NUAI,Committee,新闻,Uncategorized(0)arXiv:2507.15778v2 Announce Type: replace Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become an effective...
