English
Related papers

Related papers: Token-Level Entropy Reveals Demographic Disparitie…

200 papers

Through the use of first name substitution experiments, prior research has demonstrated the tendency of social commonsense reasoning models to systematically exhibit social biases along the dimensions of race, ethnicity, and gender (An et…

Computation and Language · Computer Science 2023-05-29 Haozhe An , Rachel Rudinger

Large language models (LLMs) are increasingly used for automated text annotation in tasks ranging from academic research to content moderation and hiring. Across 19 LLMs and two experiments totaling more than 4 million annotation judgments,…

Computation and Language · Computer Science 2026-03-17 Petter Törnberg

All AI models are susceptible to learning biases in data that they are trained on. For generative dialogue models, being trained on real human conversations containing unbalanced gender and race/ethnicity references can lead to models that…

Computation and Language · Computer Science 2021-09-09 Eric Michael Smith , Adina Williams

Name-based gender prediction has traditionally categorized individuals as either female or male based on their names, using a binary classification system. That binary approach can be problematic in the cases of gender-neutral names that do…

Computation and Language · Computer Science 2024-07-09 Zhiwen You , HaeJin Lee , Shubhanshu Mishra , Sullam Jeoung , Apratim Mishra , Jinseok Kim , Jana Diesner

We introduce a simple, yet novel entropy-based framework to drive token efficiency in large language models during reasoning tasks. Our approach uses Shannon entropy from token-level logprobs as a confidence signal to enable early stopping,…

Machine Learning · Computer Science 2025-10-29 Aman Sharma , Paras Chopra

We investigate how LLMs encode sociodemographic attributes of human conversational partners inferred from indirect cues such as names and occupations. We show that LLMs develop linear representations of user demographics within activation…

Artificial Intelligence · Computer Science 2025-12-12 Paul Bouchaud , Pedro Ramaciotti

We use a dataset of U.S. first names with labels based on predominant gender and racial group to examine the effect of training corpus frequency on tokenization, contextualization, similarity to initial representation, and bias in BERT,…

Computers and Society · Computer Science 2021-10-05 Robert Wolfe , Aylin Caliskan

Large language models (LLMs) increasingly operate in high-stakes settings including healthcare and medicine, where demographic attributes such as race and ethnicity may be explicitly stated or implicitly inferred from text. However,…

Computation and Language · Computer Science 2026-01-21 Shiyue Hu , Ruizhe Li , Yanjun Gao

Language models cannot be random. This paper introduces Entropic Deviation (ED), the normalised KL divergence between a model's token distribution and the uniform distribution, and measures it systematically across 31,200 generations…

Computation and Language · Computer Science 2026-04-28 Jarosław Hryszko

Subword tokenization is a key part of many NLP pipelines. However, little is known about why some tokenizer and hyperparameter combinations lead to better downstream model performance than others. We propose that good tokenizers lead to…

Computation and Language · Computer Science 2023-06-30 Vilém Zouhar , Clara Meister , Juan Luis Gastaldi , Li Du , Mrinmaya Sachan , Ryan Cotterell

Deep neural networks (DNNs) have the capacity to fit extremely noisy labels nonetheless they tend to learn data with clean labels first and then memorize those with noisy labels. We examine this behavior in light of the Shannon entropy of…

Machine Learning · Computer Science 2021-04-28 Hao Wu , Jiangchao Yao , Jiajie Wang , Yinru Chen , Ya Zhang , Yanfeng Wang

Prior work has explored directly regularizing the output distributions of probabilistic models to alleviate peaky (i.e. over-confident) predictions, a common sign of overfitting. This class of techniques, of which label smoothing is one,…

Computation and Language · Computer Science 2020-05-13 Clara Meister , Elizabeth Salesky , Ryan Cotterell

Conversational data are central to the study of interaction dynamics and social structures across psychological research. However, constructing structured social networks from unstructured conversational data remains a major methodological…

Applications · Statistics 2026-03-04 Gwanghee Kim , Ick Hoon Jin , Minjeong Jeon

Gender and race inferred from an individual's name are a notable source of stereotypes and biases that subtly influence social interactions. Abundant evidence from human experiments has revealed the preferential treatment that one receives…

Computers and Society · Computer Science 2024-04-24 Yumou Wei , Paulo F. Carvalho , John Stamper

Large language models are increasingly used as writing tools and pedagogical resources in English for Academic Purposes, but it remains unclear whether they encode culturally differentiated representations when generating academic text.…

Computation and Language · Computer Science 2026-04-15 Paul Jackson , Ruizhe Li , Elspeth Edelstein

As language models continue to be integrated into applications of personal and societal relevance, ensuring these models' trustworthiness is crucial, particularly with respect to producing consistent outputs regardless of sensitive…

Computation and Language · Computer Science 2023-06-05 Sullam Jeoung , Jana Diesner , Halil Kilicoglu

Large language models (LLMs) are becoming pervasive in everyday life, yet their propensity to reproduce biases inherited from training data remains a pressing concern. Prior investigations into bias in LLMs have focused on the association…

Computation and Language · Computer Science 2024-04-29 Messi H. J. Lee , Jacob M. Montgomery , Calvin K. Lai

Large language models (LLMs) have demonstrated remarkable capabilities in simulating human behaviour and social intelligence. However, they risk perpetuating societal biases, especially when demographic information is involved. We introduce…

Computers and Society · Computer Science 2025-06-11 Bryan Chen Zhengyu Tan , Roy Ka-Wei Lee

Allometric growth is found in many tagging systems online. That is, the number of new tags (T) is a power law function of the active population (P), or T P^gamma (gamma!=1). According to previous studies, it is the heterogeneity in…

Information Retrieval · Computer Science 2012-01-31 Lingfei Wu , Chengjun Wang

Masked language modeling has become a standard pretraining objective for training encoder-based language models. In this approach, certain tokens in the input are masked, and the model learns to predict them using the surrounding context.…

Artificial Intelligence · Computer Science 2026-05-28 Gokul Srinivasagan , Kai Hartung , Munir Georges
‹ Prev 1 2 3 10 Next ›