Progressive Sparse Attention: Algorithm and System Co-design for Efficient Attention in LLM Serving — mentioned by JPLeRouzic

Wed, 30 Sep 2026 07:49:44 GMT

I have read somewhere that Transformer architecture has a quadratic cost [0] (which explains the high costs associated with LLMs and the difficulty for constant improvement without state size pockets). For what I understand PSSA belongs to a line of research for LLMs with scalable architecture because you don't need to load the full KV in memory to generate a single token: [0] https://aclanthology.org/2023.findings-emnlp.936/ https://arxiv.org/abs/2503.00392 https://papers.nips.cc/paper_files/paper/2023/hash/6ceefa7b1...

Paper and discussion history. A mention is not an endorsement.

Progressive Sparse Attention: Algorithm and System Co-design for Efficient Attention in LLM Serving · HackerPapers discussions

Progressive Sparse Attention: Algorithm and System Co-design for Efficient Attention in LLM Serving — mentioned by JPLeRouzic