8 papers
Algebraic Decomposition Theory for Transformer Length Generalization
Andy Yang, Blerta Veseli, Corentin Barloy +5
Transformer-based language models are known to sometimes generalize to sequences longer than seen during training, but we lack a precise characterization of which tasks admit lengt…
Length Generalization Bounds for Transformers
Andy Yang, Pascal BergsträÃer, Georg Zetzsche +2
Length generalization is a key property of a learning algorithm that enables it to make correct predictions on inputs of any length, given finite training data. To provide such a g…
Probability Distributions Computed by Autoregressive Transformers
Andy Yang, Anej Svete, Jiaoda Li +4
Most expressivity results for transformers treat them as language recognizers -- devices that accept or reject strings -- rather than as they are used in practice: as language mode…
Knee-Deep in C-RASP: A Transformer Depth Hierarchy
Andy Yang, Michaël Cadilhac, David Chiang
It has been observed that transformers with greater depth (that is, more layers) have more capabilities, but can we establish formally which capabilities are gained? We answer this…
The Transformer Cookbook
Andy Yang, Christopher Watson, Anton Xue +6
We present the transformer cookbook: a collection of techniques for directly encoding algorithms into a transformer's parameters. This work addresses the steep learning curve of su…
Simulating Hard Attention Using Soft Attention
Andy Yang, Lena Strobl, David Chiang +1
We study conditions under which transformers using soft attention can simulate hard attention, that is, effectively focus all attention on a subset of positions. First, we examine…