{"slug": "lokiformer-locality-aware-attention-with-decoupled-knowledge-memory-for-large", "title": "LoKiFormer: Locality-aware Attention with Decoupled Knowledge Memory for Efficient Large Language Model Pretraining", "summary": "Researchers propose LoKiFormer, a novel large language model architecture that augments the standard decoder with Local Fusion Attention and a Knowledge Memory Module to address inefficiencies in pretraining. According to the arXiv paper (2608.12419v1), LoKiFormer converges 1.33x faster in pre-training than baseline models, demonstrating improved efficiency in integrating local and global information.", "body_md": "arXiv:2608.12419v1 Announce Type: new\nAbstract: Large language models (LLMs) have achieved remarkable breakthroughs across various applications. However, their architectures remain inefficient in pretraining due to two main limitations: (i) self-attention lacks an explicit inductive bias for locality, leading to redundant modeling of sequence-internal local information; (ii) mixture-of-experts (MoE) implicitly couples knowledge storage with computational pathways, hindering flexible access to sequence-external global knowledge. To overcome these limitations, we propose LoKiFormer, a novel LLM architecture that augments the standard decoder with two dedicated modules: 1) Local Fusion Attention (LFA), which incorporates a convolutional fusion to attention, explicitly capturing local patterns and allowing the attention to operate on more informative representations; 2) Knowledge Memory Module (KMM), which introduces a parametric key-value memory that explicitly stores global knowledge in addressable slots, decoupling storage from computation and enabling direct knowledge retrieval. Together, these modules enable LoKiFormer to achieve more efficient and effective integration of information at both levels. Experimental results show that LoKiFormer converges 1.33x faster in pre-training than baseline models, underscoring its superiority over existing LLM architectures.", "url": "https://wpnews.pro/news/lokiformer-locality-aware-attention-with-decoupled-knowledge-memory-for-large", "canonical_source": "https://arxiv.org/abs/2608.12419", "published_at": "2026-08-14 04:00:00+00:00", "updated_at": "2026-08-14 04:15:25.706569+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "machine-learning"], "entities": ["LoKiFormer", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/lokiformer-locality-aware-attention-with-decoupled-knowledge-memory-for-large", "markdown": "https://wpnews.pro/news/lokiformer-locality-aware-attention-with-decoupled-knowledge-memory-for-large.md", "text": "https://wpnews.pro/news/lokiformer-locality-aware-attention-with-decoupled-knowledge-memory-for-large.txt", "jsonld": "https://wpnews.pro/news/lokiformer-locality-aware-attention-with-decoupled-knowledge-memory-for-large.jsonld"}}