• inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.
• embedding — A list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.

Injecting only the relevant ontology slice per document cuts prompt catalog overhead ~94% while pushing search recall from about 70% to 95%, and the conflict guard shows how to stop embeddings from silently merging two people with the same name.

• articleNexus: Depth-Adaptive KV-Cache Splicing and Retrieval-Decoupled Tool Routing for Agentic LLMs on Unified MemoryMustafa Arslan
• articleMergeable Model-Side Aggregation States for Long-Context Language ModelsDachuan Song, Junyu Yin, Zechen Hu, Xuan Wang
• articleDocAtlas: Long-Document Understanding as Mutable-State InteractionHongchen Wei, Yuanzhe Wang, Bei Liu, Yifan Yang, Qi Dai, Kai Qiu, Yunsheng Li, Dongdong Chen, Chong Luo, Zhenzhong Chen, Baining Guo

Checking sign-in…

Loading comments…