paper-with-me

Papers

High probability or low information? The probability–quality paradox in language generation

2022-05-01 · ACL 2022 5 · Clara Meister, Gian Wiher, Tiago Pimentel, Ryan Cotterell

When generating natural language from neural probabilistic models, high probability does not always coincide with high quality. Rather, mode-seeking decoding methods can lead to incredibly unnatural language, while stochastic methods produce text perceived as much more human-like. In this note, we offer an explanation for this phenomenon by analyzing language as a means of communication in the information-theoretic sense. We posit that human-like language usually contains an expected amount of information—quantified as negative log-probability—and that language with substantially more (or less) information is undesirable. We provide preliminary empirical evidence for this hypothesis using quality ratings for both human and machine-generated text, covering multiple tasks and common decoding schemes.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Text Generation

Similar Papers 제목 키워드 기반

On the probability-quality paradox in language generation

2022-03-31 · Clara Meister, Gian Wiher, Tiago Pimentel, Ryan Cotterell

When generating natural language from neural probabilistic models, high probability does not always coincide with high quality: It has often been observed that mode-seeking decoding methods, i.e., those that produce high…

Text Generation

Causal Confirmation Measures: From Simpson's Paradox to COVID-19

2023-02-03 · Chenguang Lu

When we compare the influences of two causes on an outcome, if the conclusion from every group is against that from the conflation, we think there is Simpson's Paradox. The Existing Causal Inference Theory (ECIT) can mak…

Causal Inference

Towards a Measure Theory of Semantic Information

2025-08-01 · George M. Coghill arxiv

A classic account of the quantification of semantic information is that of Bar-Hiller and Carnap. Their account proposes an inverse relation between the informativeness of a statement and its probability. However, their …

On Nicod's Condition, Rules of Induction and the Raven Paradox

2013-07-12 · Hadi Mohasel Afshar, Peter Sunehag

Philosophers writing about the ravens paradox often note that Nicod's Condition (NC) holds given some set of background information, and fails to hold against others, but rarely go any further. That is, it is usually not…

Channels' Confirmation and Predictions' Confirmation: from the Medical Test to the Raven Paradox

2020-01-17 · Chenguang Lu

After long arguments between positivism and falsificationism, the verification of universal hypotheses was replaced with the confirmation of uncertain major premises. Unfortunately, Hemple discovered the Raven Paradox (R…