architecture · first seen 7 Oct 2026
hybrid attention models
Also: hybrid models, hybrid modeling
Language models combining full attention with efficient attention variants such as sliding-window or linear attention.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Mechanics of Long-Context Hybrid Models Part 1.1: From Hybrid Attention to Hybrid PositionPapers · 7 Oct 2026 · Xiaoran Liu, Ziwei He, Xipeng Qiu
- Non-Interfering Asymmetric AdamW Updates for Constrained Exploration of Equifinality in Hybrid ModelsPapers · 3 Nov 2026 · Ekaterina Gikalo, Zavud Baghirov, Christian Reimers, Markus Reichstein, Martin J…
- SinkState: On-the-Fly SVD of Recurrent States for Efficient Hybrid LLM ServingPapers · 10 Nov 2026 · Haolin Jiang, Weihang Li, Zhongzhu Zhou, Jinwen Pan, Santosh Pandey, Hang Liu
Latest sources
- SinkState: On-the-Fly SVD of Recurrent States for Efficient Hybrid LLM ServingPapers · 10 Nov 2026 · Haolin Jiang, Weihang Li, Zhongzhu Zhou, Jinwen Pan, Santosh Pandey, Hang Liu
- Non-Interfering Asymmetric AdamW Updates for Constrained Exploration of Equifinality in Hybrid ModelsPapers · 3 Nov 2026 · Ekaterina Gikalo, Zavud Baghirov, Christian Reimers, Markus Reichstein, Martin J…
- Mechanics of Long-Context Hybrid Models Part 1.1: From Hybrid Attention to Hybrid PositionPapers · 7 Oct 2026 · Xiaoran Liu, Ziwei He, Xipeng Qiu