Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Length Generalization for Transformers via Compression
Georg Zetzsche, Hongjian Jiang, Andy Yang +4
Recent advancements in transformer length generalization theory enable us to reliably predict when a transformer can learn to solve a task. In particular, the C-RASP hypothesis (a…
cs.LG2026
Length Generalization Bounds for Transformers
Andy Yang, Pascal Bergsträßer, 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…