arXiv Computation and Language
surprisal is Not a Theory
Read the original on arXiv Computation and Language →The article argues that Surprisal Theory, often presented as a computational-level explanation, is not a theory in its own right. It contends that using large language model (LLM) surprisals without considering the underlying representational and algorithmic choices obscures the theory’s commitments. The authors demonstrate through three analyses that algorithm and architecture significantly influence language model probabilities, urging researchers to reassess treating LLM surprisals as interchangeable.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.