While Large Language Models (LLMs) advance 3D indoor scene synthesis, current pipelines fail to retain user-specific pre…
为什么值得关注：属于技术进展，可能扩散到其他领域
AI agents can each make locally valid decisions yet jointly produce an invalid result.We call this the global coherence…
为什么值得关注：属于政府/政策信息，对教育或研究方向可能产生影响
Large language model (LLM) agents increasingly combine reasoning, tool use, and action, but most evidence comes from epi…
为什么值得关注：属于技术进展，可能扩散到其他领域
Recent years have witnessed the rapid adoption of reinforcement learning (RL) in large language model (LLM) post-trainin…
为什么值得关注：属于政府/政策信息，对教育或研究方向可能产生影响
Schrödinger bridge (SB) learns stochastic transport between prescribed initial and target distributions.When the initial…
为什么值得关注：属于政府/政策信息，对教育或研究方向可能产生影响
When a SCADA alarm is raised in a water distribution network, operators must decide quickly whether it reflects a cybera…
为什么值得关注：属于企业动态，反映产业投入方向
As Large Language Models (LLMs) increasingly serve as foundational reasoning engines, their tendency to hallucinate rema…
为什么值得关注：属于技术进展，可能扩散到其他领域
Memory-augmented large language models must decide which memories to retain, and recent systems do so by estimating each…
为什么值得关注：属于政府/政策信息，对教育或研究方向可能产生影响
The study of ceramic materials constitutes a cornerstone of archaeological research, yet the post-production workflow fo…
为什么值得关注：属于技术进展，可能扩散到其他领域
Privacy-preserving machine learning presents significant deployment challenges on the cloud for intelligent systems with…
为什么值得关注：属于政府/政策信息，对教育或研究方向可能产生影响
As robotic hardware and learning methods advance, humanoids need tools to perform tasks beyond their inherent physical l…
为什么值得关注：属于技术进展，可能扩散到其他领域
Despite progress in vision-language models, 3D spatial reasoning from 2D images remains challenging.Text-based methods d…
为什么值得关注：属于企业动态，反映产业投入方向
Data selection is critical for training large language models on massive and heterogeneous corpora. Meta-learning for Tr…
A comparative explainability framework is presented to audit DeBERTa-v3 under zero-shot classification of medical abstra…
On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supe…
Real-world enterprise data science and analytics workflows require reasoning across dozens of tables, performing statist…
We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric…
Unsupervised anomaly detection (UAD) methods for brain MRI are ranked by a single score, yet that score rests on choices…
Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supe…
Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-inten…