FlashAttention-2: Faster attention with better parallelism and work partitioningtogether.ai·2 pts·espadrine·0
Lightweight Cryptography Standardization Process: NIST Selects Asconcsrc.nist.gov·33 pts·espadrine·10
AudioLM: A Language Modeling Approach to Audio Generationgoogle-research.github.io·58 pts·espadrine·6
Minerva: Solving Quantitative Reasoning Problems with Language Modelsminerva-demo.github.io·44 pts·espadrine·1
Rethinking price plans to support a fast growing and changing customer basemedium.com·1 pts·espadrine·0
A connectomic study of a petascale fragment of human cerebral cortex [pdf]biorxiv.org·1 pts·espadrine·0