Transformers need glasses! Information over-squashing in language tasks

Federico Barbero, Andrea Banino, Steven Kapturowski, Dharshan Kumaran, João Guilherme Madeira Araújo, Alex Vitvitskyi, Razvan Pascanu, Petar Veličković

Keywords: Transformers, Over-squashing, Large Language Models

Abstract: We study how information propagates in decoder-only Transformers, which are the architectural foundation of most existing frontier large language models (LLMs). We rely on a theoretical signal propagation analysis---specifically, we analyse the representations of the last token in the final layer of the Transformer, as this is the representation used for next-token prediction. Our analysis reveals a representational collapse phenomenon: we prove that certain distinct pairs of inputs to the Transformer can yield arbitrarily close representations in the final token. This effect is exacerbated by the low-precision floating-point formats frequently used in modern LLMs. As a result, the model is provably unable to respond to these sequences in different ways---leading to errors in, e.g., tasks involving counting or copying. Further, we show that decoder-only Transformer language models can lose sensitivity to specific tokens in the input, which relates to the well-known phenomenon of over-squashing in graph neural networks. We provide empirical evidence supporting our claims on contemporary LLMs. Our theory points to simple solutions towards ameliorating these issues.

Primary Area: Natural language processing

Submission Number: 18585