Tencent releases SAPS: sparse attention that learns KV block selection end-to-end
teortaxesTex · x · 2026-09-15
Tencent has released a new sparse attention method, Simple Attention Sparsification (SAPS), on Hugging Face.
The key idea: instead of relying on handcrafted heuristics for which KV blocks to attend to, SAPS lets the language modeling loss directly optimize block selection per query — making sparse attention end-to-end learnable and potentially cutting long-context inference cost.
Related event: Tencent Open-Sources SAPS Sparse Attention Method for Qwen3(2 posts)→
More from Research
- Yandex open-sources its search AI answer model, squeezing 40% more answers from same compute — teortaxesTex · 2026-09-15
- AI face-preference study goes megaviral with 450,000 completers, v2.0 released — SpencrGreenberg · 2026-09-15
- Scholars blast arXiv for losing its way as a preprint server over AI crackdowns — RexDouglass · 2026-09-15
- BuzzASR releases 102 monolingual ASR models, beating Whisper on 77 languages with SOTA on 27 — LChoshen · 2026-09-15
- ICIP 2026 plenary: computational imaging shifts from explicit priors to learned operators — prof_kamilov · 2026-09-15
- Denny Zhou calls for arXiv submission fees as top ML conferences move that way — MengdiWang10 · 2026-09-15