Do we need Attention? - Linear RNNs and State Space Models (SSMs) for NLP
Why Hacker News recommends it
HN commenters recommend it for building intuition about state space models and the tradeoffs between transformer and SSM layers.
What it is
Sasha Rush surveys linear RNNs and state space models as replacements for attention in transformer-style models, in a talk for MLSys. It covers the foundational preliminaries and predates Mamba.
Recommended by 3 HN commenters across 3 different months, 2024 to 2024.
What Hacker News said
Strong rec a_wild_dandan · 2024-03 · on “Jamba: Production-grade Mamba-based AI model”
To those curious about the tradeoffs between transformer and state space model layers, I highly recommend Sasha Rush's video on it: [this video]
Recommends anomaly72 · 2024-01 · on “MoE-Mamba: Efficient Selective State Space Models with Mixture of Experts”
Would recommend Sasha Rush's lecture [this video] , it provides the intuition for state space models via linear RNNs
Recommends srush · 2024-02 · on “The Annotated S4”
Hi! Blog author. This was an attempt a couple years ago to understand and write about this paper in a detailed way. Here is a video going through this topic as well: [this video] Section 2 of the blog post is no longer very relevant. A lot of advances (DSS, S4D) simplified that part of the process. Arguably also this all should be updated for Mamba (same authors).