Temperature-scaling surprisal estimates improve fit to human reading times -- but does it do so for the "right reasons"?

• URL: http://arxiv.org/abs/2311.09325v2
• Date: Wed, 3 Jul 2024 16:12:32 GMT
• Title: Temperature-scaling surprisal estimates improve fit to human reading times -- but does it do so for the "right reasons"?
• Authors: Tong Liu, Iza Škrjanec, Vera Demberg,
• Abstract summary: We show that calibration of large language models typically improves with model size. We find that temperature-scaling probabilities lead to a systematically better fit to reading times.
• Score: 15.773775387121097
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: A wide body of evidence shows that human language processing difficulty is predicted by the information-theoretic measure surprisal, a word's negative log probability in context. However, it is still unclear how to best estimate these probabilities needed for predicting human processing difficulty -- while a long-standing belief held that models with lower perplexity would provide more accurate estimates of word predictability, and therefore lead to better reading time predictions, recent work has shown that for very large models, psycholinguistic predictive power decreases. One reason could be that language models might be more confident of their predictions than humans, because they have had exposure to several magnitudes more data. In this paper, we test what effect temperature-scaling of large language model (LLM) predictions has on surprisal estimates and their predictive power of reading times of English texts. Firstly, we show that calibration of large language models typically improves with model size, i.e. poorer calibration cannot account for poorer fit to reading times. Secondly, we find that temperature-scaling probabilities lead to a systematically better fit to reading times (up to 89% improvement in delta log likelihood), across several reading time corpora. Finally, we show that this improvement in fit is chiefly driven by words that are composed of multiple subword tokens.

Related papers

• XForecast: Evaluating Natural Language Explanations for Time Series Forecasting [72.57427992446698]
  Time series forecasting aids decision-making, especially for stakeholders who rely on accurate predictions. Traditional explainable AI (XAI) methods, which underline feature or temporal importance, often require expert knowledge. evaluating forecast NLEs is difficult due to the complex causal relationships in time series data.
  arXiv  Detail & Related papers  (2024-10-18T05:16:39Z)
• Frequency Explains the Inverse Correlation of Large Language Models' Size, Training Data Amount, and Surprisal's Fit to Reading Times [15.738530737312335]
  Recent studies have shown that as Transformer-based language models become larger and are trained on very large amounts of data, the fit of their surprisal estimates to naturalistic human reading times degrades. This paper presents a series of analyses showing that word frequency is a key explanatory factor underlying these two trends. The results indicate that Transformer-based language models' surprisal estimates diverge from human-like expectations due to the superhumanly complex associations they learn for predicting rare words.
  arXiv  Detail & Related papers  (2024-02-03T20:22:54Z)
• Are We Falling in a Middle-Intelligence Trap? An Analysis and Mitigation of the Reversal Curse [73.65112477688353]
  Recent studies have highlighted a phenomenon in large language models known as "the reversal curse" We contend that the reversal curse is partially a result of specific model training objectives. We propose a novel training method, BI Casual language modeling Optimization (BICO), designed to mitigate the reversal curse.
  arXiv  Detail & Related papers  (2023-11-13T17:01:12Z)
• Humans and language models diverge when predicting repeating text [52.03471802608112]
  We present a scenario in which the performance of humans and LMs diverges. Human and GPT-2 LM predictions are strongly aligned in the first presentation of a text span, but their performance quickly diverges when memory begins to play a role. We hope that this scenario will spur future work in bringing LMs closer to human behavior.
  arXiv  Detail & Related papers  (2023-10-10T08:24:28Z)
• Testing the Predictions of Surprisal Theory in 11 Languages [77.45204595614]
  We investigate the relationship between surprisal and reading times in eleven different languages. By focusing on a more diverse set of languages, we argue that these results offer the most robust link to-date between information theory and incremental language processing across languages.
  arXiv  Detail & Related papers  (2023-07-07T15:37:50Z)
• Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times? [9.909170013118775]
  The propensity of larger Transformer-based models to'memorize' sequences during training makes their surprisal estimates diverge from humanlike expectations. These results suggest that the propensity of larger Transformer-based models to'memorize' sequences during training makes their surprisal estimates diverge from humanlike expectations.
  arXiv  Detail & Related papers  (2022-12-23T03:57:54Z)
• Pathologies of Pre-trained Language Models in Few-shot Fine-tuning [50.3686606679048]
  We show that pre-trained language models with few examples show strong prediction bias across labels. Although few-shot fine-tuning can mitigate the prediction bias, our analysis shows models gain performance improvement by capturing non-task-related features. These observations alert that pursuing model performance with fewer examples may incur pathological prediction behavior.
  arXiv  Detail & Related papers  (2022-04-17T15:55:18Z)
• You Can Do Better! If You Elaborate the Reason When Making Prediction [13.658942796267015]
  This paper proposes a novel neural predictive framework coupled with large pre-trained language models to make a prediction and generate its corresponding explanation simultaneously. We conducted a preliminary empirical study on Chinese medical multiple-choice question answering, English natural language inference and commonsense question answering tasks. The proposed method also achieves improved prediction accuracy on three datasets, which indicates that making predictions can benefit from generating the explanation in the decision process.
  arXiv  Detail & Related papers  (2021-03-27T14:55:19Z)
• Are Some Words Worth More than Others? [3.5598388686985354]
  We propose two new intrinsic evaluation measures within the framework of a simple word prediction task. We evaluate several commonly-used large English language models using our proposed metrics.
  arXiv  Detail & Related papers  (2020-10-12T23:12:11Z)
• Are Visual Explanations Useful? A Case Study in Model-in-the-Loop Prediction [49.254162397086006]
  We study explanations based on visual saliency in an image-based age prediction task. We find that presenting model predictions improves human accuracy. However, explanations of various kinds fail to significantly alter human accuracy or trust in the model.
  arXiv  Detail & Related papers  (2020-07-23T20:39:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.