paper-with-me

홈 › Papers

The human unlikeness of neural language models in next-word prediction

2020-07-01 · WS 2020 7 · Cass Jacobs, ra L., Arya D. McCarthy

The training objective of unidirectional language models (LMs) is similar to a psycholinguistic benchmark known as the cloze task, which measures next-word predictability. However, LMs lack the rich set of experiences that people do, and humans can be highly creative. To assess human parity in these models{'} training objective, we compare the predictions of three neural language models to those of human participants in a freely available behavioral dataset (Luke {\&} Christianson, 2016). Our results show that while neural models show a close correspondence to human productions, they nevertheless assign insufficient probability to how often speakers guess upcoming words, especially for open-class content words.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Cloze Distillation: Improving Neural Language Models with Human Next-Word Prediction

2020-11-01 · CONLL 2020 · Tiwalayo Eisape, Noga Zaslavsky, Roger Levy

Contemporary autoregressive language models (LMs) trained purely on corpus data have been shown to capture numerous features of human incremental processing. However, past work has also suggested dissociations between co…

Humans and language models diverge when predicting repeating text

2023-10-10 · Aditya R. Vaidya, Javier Turek, Alexander G. Huth

Language models that are trained on the next-word prediction task have been shown to accurately model human behavior in word prediction and reading speed. In contrast with these findings, we present a scenario in which t…

In-Context Learning

Learning to vary: Teaching LMs to reproduce human linguistic variability in next-word prediction

2025-09-22 · Tobias Groot, Salo Lacunes, Evgenia Ilia arxiv

Natural language generation (NLG) tasks are often subject to inherent variability; e.g. predicting the next word given a context has multiple valid responses, evident when asking multiple humans to complete the task. Whi…

Next word prediction based on the N-gram model for Kurdish Sorani and Kurmanji

2020-07-27 · Hozan K. Hamarashid, Soran A. Saeed, Tarik A. Rashid

Next word prediction is an input technology that simplifies the process of typing by suggesting the next word to a user to select, as typing in a conversation consumes time. A few previous studies have focused on the Kur…

Prediction

To model human linguistic prediction, make LLMs less superhuman

2025-10-01 · Byung-Doh Oh, Tal Linzen arxiv

When we read, we make predictions about upcoming words; these predictions influence our reading behavior. The success of large language models (LLMs), which, like humans, make predictions about upcoming words, has motiva…