“There's this nice new paper ("Simplifying Transformer Blocks" https://arxiv.org/abs/2311.01906) about getting rid of all the unnecessary parts of a transformer. On their dataset, they manage to get rid of the layernorm, the skip connection, the value matrix, and the projection matrix.”
Simplifying Transformer Blocks
A simple design recipe for deep Transformers is to compose identical building blocks. But standard transformer blocks are far from simple, interweaving attention and MLP sub-blocks with skip connections & normalisation layers in precise arrangements. This complexity leads to brittle architectures, where seemingly minor changes can significantly reduce training speed, or render models untrainable. In this work, we ask to what extent the standard transformer block can be simplified? Combining signal propagation theory and empirical observations, we motivate modifications that allow many block components to be removed with no loss of training speed, including skip connections, projection or value parameters, sequential sub-blocks and normalisation layers. In experiments on both autoregressive decoder-only and BERT encoder-only models, our simplified transformers emulate the per-update training speed and performance of standard transformers, while enjoying 15% faster training throughput, and using 15% fewer parameters.
候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。
HN 书评
推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及“Just skimmed so far and didn't see any reference to the Simplified Transformer block of https://arxiv.org/abs/2311.01906 (and it seems they also left out grouped query attention, too, as pointed out by another comment). While lazy me wants them to explain how their approach compares to these approaches, it looks like their exposition is pretty clear (qui…”
- Simplifying Transformer Blocks ↗lordswork · 2023-11-09 · 1 分
- Simplifying Transformer Blocks ↗georgehill · 2023-11-28 · 2 分