SR-KI: Scalable and Real-Time Knowledge Integration into LLMs via Supervised Attention
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (41) · 34486-34494
DOI: 10.1609/aaai.v40i41.40747 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Bohan Yu und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
This paper proposes SR-KI, a novel approach for integrating real-time and large-scale structured knowledge bases (KBs) into large language models (LLMs). SR-KI begins by encoding KBs into key-value pairs using a pretrained encoder, and injects them into LLMs' KV cache. Building on this representation, we employ a two-stage training paradigm: first locating a dedicated retrieval layer within the LLM, and then applying an attention-based loss at this layer to explicitly supervise attention toward relevant KB entries. Unlike traditional retrieval-augmented generation methods that rely heavily on the performance of external retrievers and multi-stage pipelines, SR-KI supports end-to-end inference by performing retrieval entirely within the model’s latent space. This design enables efficient compression of injected knowledge and facilitates dynamic knowledge updates. Comprehensive experiments demonstrate that SR-KI enables the integration of up to 40K KBs into a 7B LLM on a single A100 40GB GPU, and achieves strong retrieval performance—maintaining over 98% Recall@10 on the best-performing task and exceeding 88% on average across all tasks. Task performance on question answering and KB ID generation also demonstrates that SR-KI maintains strong performance while achieving up to 99.75% compression of the injected KBs.
Zitieren
APA
Yu, B., Huang, W., & Liu, K. (2026). SR-KI: Scalable and Real-Time Knowledge Integration into LLMs via Supervised Attention. Proceedings of the AAAI Conference on Artificial Intelligence, 40(41), 34486-34494. https://doi.org/10.1609/aaai.v40i41.40747
Vancouver
Yu B, Huang W, Liu K. SR-KI: Scalable and Real-Time Knowledge Integration into LLMs via Supervised Attention. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(41):34486-34494. doi:10.1609/aaai.v40i41.40747
BibTeX
@article{Yu2026,
author = {Yu, Bohan and Huang, Wei and Liu, Kang},
title = {SR-KI: Scalable and Real-Time Knowledge Integration into LLMs via Supervised Attention},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {41},
pages = {34486--34494},
doi = {10.1609/aaai.v40i41.40747}
}
EUVIMED