-
AboutMe: Using Self-Descriptions in Webpages to Document the Effects of English Pretraining Data Filters
Authors:
Li Lucy,
Suchin Gururangan,
Luca Soldaini,
Emma Strubell,
David Bamman,
Lauren F. Klein,
Jesse Dodge
Abstract:
Large language models' (LLMs) abilities are drawn from their pretraining data, and model development begins with data curation. However, decisions around what data is retained or removed during this initial stage are under-scrutinized. In our work, we ground web text, which is a popular pretraining data source, to its social and geographic contexts. We create a new dataset of 10.3 million self-des…
▽ More
Large language models' (LLMs) abilities are drawn from their pretraining data, and model development begins with data curation. However, decisions around what data is retained or removed during this initial stage are under-scrutinized. In our work, we ground web text, which is a popular pretraining data source, to its social and geographic contexts. We create a new dataset of 10.3 million self-descriptions of website creators, and extract information about who they are and where they are from: their topical interests, social roles, and geographic affiliations. Then, we conduct the first study investigating how ten "quality" and English language identification (langID) filters affect webpages that vary along these social dimensions. Our experiments illuminate a range of implicit preferences in data curation: we show that some quality classifiers act like topical domain filters, and langID can overlook English content from some regions of the world. Overall, we hope that our work will encourage a new line of research on pretraining data curation practices and its social implications.
△ Less
Submitted 20 June, 2024; v1 submitted 12 January, 2024;
originally announced January 2024.
-
Social Meme-ing: Measuring Linguistic Variation in Memes
Authors:
Naitian Zhou,
David Jurgens,
David Bamman
Abstract:
Much work in the space of NLP has used computational methods to explore sociolinguistic variation in text. In this paper, we argue that memes, as multimodal forms of language comprised of visual templates and text, also exhibit meaningful social variation. We construct a computational pipeline to cluster individual instances of memes into templates and semantic variables, taking advantage of their…
▽ More
Much work in the space of NLP has used computational methods to explore sociolinguistic variation in text. In this paper, we argue that memes, as multimodal forms of language comprised of visual templates and text, also exhibit meaningful social variation. We construct a computational pipeline to cluster individual instances of memes into templates and semantic variables, taking advantage of their multimodal structure in doing so. We apply this method to a large collection of meme images from Reddit and make available the resulting \textsc{SemanticMemes} dataset of 3.8M images clustered by their semantic function. We use these clusters to analyze linguistic variation in memes, discovering not only that socially meaningful variation in meme usage exists between subreddits, but that patterns of meme innovation and acculturation within these communities align with previous findings on written language.
△ Less
Submitted 15 November, 2023;
originally announced November 2023.
-
Grounding Characters and Places in Narrative Texts
Authors:
Sandeep Soni,
Amanpreet Sihra,
Elizabeth F. Evans,
Matthew Wilkens,
David Bamman
Abstract:
Tracking characters and locations throughout a story can help improve the understanding of its plot structure. Prior research has analyzed characters and locations from text independently without grounding characters to their locations in narrative time. Here, we address this gap by proposing a new spatial relationship categorization task. The objective of the task is to assign a spatial relations…
▽ More
Tracking characters and locations throughout a story can help improve the understanding of its plot structure. Prior research has analyzed characters and locations from text independently without grounding characters to their locations in narrative time. Here, we address this gap by proposing a new spatial relationship categorization task. The objective of the task is to assign a spatial relationship category for every character and location co-mention within a window of text, taking into consideration linguistic context, narrative tense, and temporal scope. To this end, we annotate spatial relationships in approximately 2500 book excerpts and train a model using contextual embeddings as features to predict these relationships. When applied to a set of books, this model allows us to test several hypotheses on mobility and domestic space, revealing that protagonists are more mobile than non-central characters and that women as characters tend to occupy more interior space than men. Overall, our work is the first step towards joint modeling and analysis of characters and places in narrative text.
△ Less
Submitted 27 May, 2023;
originally announced May 2023.
-
Dramatic Conversation Disentanglement
Authors:
Kent K. Chang,
Danica Chen,
David Bamman
Abstract:
We present a new dataset for studying conversation disentanglement in movies and TV series. While previous work has focused on conversation disentanglement in IRC chatroom dialogues, movies and TV shows provide a space for studying complex pragmatic patterns of floor and topic change in face-to-face multi-party interactions. In this work, we draw on theoretical research in sociolinguistics, sociol…
▽ More
We present a new dataset for studying conversation disentanglement in movies and TV series. While previous work has focused on conversation disentanglement in IRC chatroom dialogues, movies and TV shows provide a space for studying complex pragmatic patterns of floor and topic change in face-to-face multi-party interactions. In this work, we draw on theoretical research in sociolinguistics, sociology, and film studies to operationalize a conversational thread (including the notion of a floor change) in dramatic texts, and use that definition to annotate a dataset of 10,033 dialogue turns (comprising 2,209 threads) from 831 movies. We compare the performance of several disentanglement models on this dramatic dataset, and apply the best-performing model to disentangle 808 movies. We see that, contrary to expectation, average thread lengths do not decrease significantly over the past 40 years, and characters portrayed by actors who are women, while underrepresented, initiate more new conversational threads relative to their speaking time.
△ Less
Submitted 26 May, 2023;
originally announced May 2023.
-
Speak, Memory: An Archaeology of Books Known to ChatGPT/GPT-4
Authors:
Kent K. Chang,
Mackenzie Cramer,
Sandeep Soni,
David Bamman
Abstract:
In this work, we carry out a data archaeology to infer books that are known to ChatGPT and GPT-4 using a name cloze membership inference query. We find that OpenAI models have memorized a wide collection of copyrighted materials, and that the degree of memorization is tied to the frequency with which passages of those books appear on the web. The ability of these models to memorize an unknown set…
▽ More
In this work, we carry out a data archaeology to infer books that are known to ChatGPT and GPT-4 using a name cloze membership inference query. We find that OpenAI models have memorized a wide collection of copyrighted materials, and that the degree of memorization is tied to the frequency with which passages of those books appear on the web. The ability of these models to memorize an unknown set of books complicates assessments of measurement validity for cultural analytics by contaminating test data; we show that models perform much better on memorized books than on non-memorized books for downstream tasks. We argue that this supports a case for open models whose training data is known.
△ Less
Submitted 20 October, 2023; v1 submitted 28 April, 2023;
originally announced May 2023.
-
Words as Gatekeepers: Measuring Discipline-specific Terms and Meanings in Scholarly Publications
Authors:
Li Lucy,
Jesse Dodge,
David Bamman,
Katherine A. Keith
Abstract:
Scholarly text is often laden with jargon, or specialized language that can facilitate efficient in-group communication within fields but hinder understanding for out-groups. In this work, we develop and validate an interpretable approach for measuring scholarly jargon from text. Expanding the scope of prior work which focuses on word types, we use word sense induction to also identify words that…
▽ More
Scholarly text is often laden with jargon, or specialized language that can facilitate efficient in-group communication within fields but hinder understanding for out-groups. In this work, we develop and validate an interpretable approach for measuring scholarly jargon from text. Expanding the scope of prior work which focuses on word types, we use word sense induction to also identify words that are widespread but overloaded with different meanings across fields. We then estimate the prevalence of these discipline-specific words and senses across hundreds of subfields, and show that word senses provide a complementary, yet unique view of jargon alongside word types. We demonstrate the utility of our metrics for science of science and computational sociolinguistics by highlighting two key social implications. First, though most fields reduce their use of jargon when writing for general-purpose venues, and some fields (e.g., biological sciences) do so less than others. Second, the direction of correlation between jargon and citation rates varies among fields, but jargon is nearly always negatively correlated with interdisciplinary impact. Broadly, our findings suggest that though multidisciplinary venues intend to cater to more general audiences, some fields' writing norms may act as barriers rather than bridges, and thus impede the dispersion of scholarly ideas.
△ Less
Submitted 22 May, 2023; v1 submitted 19 December, 2022;
originally announced December 2022.
-
Predicting Long-Term Citations from Short-Term Linguistic Influence
Authors:
Sandeep Soni,
David Bamman,
Jacob Eisenstein
Abstract:
A standard measure of the influence of a research paper is the number of times it is cited. However, papers may be cited for many reasons, and citation count offers limited information about the extent to which a paper affected the content of subsequent publications. We therefore propose a novel method to quantify linguistic influence in timestamped document collections. There are two main steps:…
▽ More
A standard measure of the influence of a research paper is the number of times it is cited. However, papers may be cited for many reasons, and citation count offers limited information about the extent to which a paper affected the content of subsequent publications. We therefore propose a novel method to quantify linguistic influence in timestamped document collections. There are two main steps: first, identify lexical and semantic changes using contextual embeddings and word frequencies; second, aggregate information about these changes into per-document influence scores by estimating a high-dimensional Hawkes process with a low-rank parameter matrix. We show that this measure of linguistic influence is predictive of $\textit{future}$ citations: the estimate of linguistic influence from the two years after a paper's publication is correlated with and predictive of its citation count in the following three years. This is demonstrated using an online evaluation with incremental temporal training/test splits, in comparison with a strong baseline that includes predictors for initial citation counts, topics, and lexical features.
△ Less
Submitted 24 October, 2022;
originally announced October 2022.
-
Discovering Differences in the Representation of People using Contextualized Semantic Axes
Authors:
Li Lucy,
Divya Tadimeti,
David Bamman
Abstract:
A common paradigm for identifying semantic differences across social and temporal contexts is the use of static word embeddings and their distances. In particular, past work has compared embeddings against "semantic axes" that represent two opposing concepts. We extend this paradigm to BERT embeddings, and construct contextualized axes that mitigate the pitfall where antonyms have neighboring repr…
▽ More
A common paradigm for identifying semantic differences across social and temporal contexts is the use of static word embeddings and their distances. In particular, past work has compared embeddings against "semantic axes" that represent two opposing concepts. We extend this paradigm to BERT embeddings, and construct contextualized axes that mitigate the pitfall where antonyms have neighboring representations. We validate and demonstrate these axes on two people-centric datasets: occupations from Wikipedia, and multi-platform discussions in extremist, men's communities over fourteen years. In both studies, contextualized semantic axes can characterize differences among instances of the same word type. In the latter study, we show that references to women and the contexts around them have become more detestable over time.
△ Less
Submitted 21 October, 2022;
originally announced October 2022.
-
Characterizing English Variation across Social Media Communities with BERT
Authors:
Li Lucy,
David Bamman
Abstract:
Much previous work characterizing language variation across Internet social groups has focused on the types of words used by these groups. We extend this type of study by employing BERT to characterize variation in the senses of words as well, analyzing two months of English comments in 474 Reddit communities. The specificity of different sense clusters to a community, combined with the specificit…
▽ More
Much previous work characterizing language variation across Internet social groups has focused on the types of words used by these groups. We extend this type of study by employing BERT to characterize variation in the senses of words as well, analyzing two months of English comments in 474 Reddit communities. The specificity of different sense clusters to a community, combined with the specificity of a community's unique word types, is used to identify cases where a social group's language deviates from the norm. We validate our metrics using user-created glossaries and draw on sociolinguistic theories to connect language variation with trends in community behavior. We find that communities with highly distinctive language are medium-sized, and their loyal and highly engaged users interact in dense networks.
△ Less
Submitted 12 February, 2021;
originally announced February 2021.
-
Latin BERT: A Contextual Language Model for Classical Philology
Authors:
David Bamman,
Patrick J. Burns
Abstract:
We present Latin BERT, a contextual language model for the Latin language, trained on 642.7 million words from a variety of sources spanning the Classical era to the 21st century. In a series of case studies, we illustrate the affordances of this language-specific model both for work in natural language processing for Latin and in using computational methods for traditional scholarship: we show th…
▽ More
We present Latin BERT, a contextual language model for the Latin language, trained on 642.7 million words from a variety of sources spanning the Classical era to the 21st century. In a series of case studies, we illustrate the affordances of this language-specific model both for work in natural language processing for Latin and in using computational methods for traditional scholarship: we show that Latin BERT achieves a new state of the art for part-of-speech tagging on all three Universal Dependency datasets for Latin and can be used for predicting missing text (including critical emendations); we create a new dataset for assessing word sense disambiguation for Latin and demonstrate that Latin BERT outperforms static word embeddings; and we show that it can be used for semantically-informed search by querying contextual nearest neighbors. We publicly release trained models to help drive future work in this space.
△ Less
Submitted 21 September, 2020;
originally announced September 2020.
-
Measuring Information Propagation in Literary Social Networks
Authors:
Matthew Sims,
David Bamman
Abstract:
We present the task of modeling information propagation in literature, in which we seek to identify pieces of information passing from character A to character B to character C, only given a description of their activity in text. We describe a new pipeline for measuring information propagation in this domain and publish a new dataset for speaker attribution, enabling the evaluation of an important…
▽ More
We present the task of modeling information propagation in literature, in which we seek to identify pieces of information passing from character A to character B to character C, only given a description of their activity in text. We describe a new pipeline for measuring information propagation in this domain and publish a new dataset for speaker attribution, enabling the evaluation of an important component of this pipeline on a wider range of literary texts than previously studied. Using this pipeline, we analyze the dynamics of information propagation in over 5,000 works of fiction, finding that information flows through characters that fill structural holes connecting different communities, and that characters who are women are depicted as filling this role much more frequently than characters who are men.
△ Less
Submitted 6 October, 2020; v1 submitted 29 April, 2020;
originally announced April 2020.
-
Breaking Speech Recognizers to Imagine Lyrics
Authors:
Jon Gillick,
David Bamman
Abstract:
We introduce a new method for generating text, and in particular song lyrics, based on the speech-like acoustic qualities of a given audio file. We repurpose a vocal source separation algorithm and an acoustic model trained to recognize isolated speech, instead inputting instrumental music or environmental sounds. Feeding the "mistakes" of the vocal separator into the recognizer, we obtain a trans…
▽ More
We introduce a new method for generating text, and in particular song lyrics, based on the speech-like acoustic qualities of a given audio file. We repurpose a vocal source separation algorithm and an acoustic model trained to recognize isolated speech, instead inputting instrumental music or environmental sounds. Feeding the "mistakes" of the vocal separator into the recognizer, we obtain a transcription of words \emph{imagined} to be spoken in the input audio. We describe the key components of our approach, present initial analysis, and discuss the potential of the method for machine-in-the-loop collaboration in creative applications.
△ Less
Submitted 15 December, 2019;
originally announced December 2019.
-
An Annotated Dataset of Coreference in English Literature
Authors:
David Bamman,
Olivia Lewke,
Anya Mansoor
Abstract:
We present in this work a new dataset of coreference annotations for works of literature in English, covering 29,103 mentions in 210,532 tokens from 100 works of fiction. This dataset differs from previous coreference datasets in containing documents whose average length (2,105.3 words) is four times longer than other benchmark datasets (463.7 for OntoNotes), and contains examples of difficult cor…
▽ More
We present in this work a new dataset of coreference annotations for works of literature in English, covering 29,103 mentions in 210,532 tokens from 100 works of fiction. This dataset differs from previous coreference datasets in containing documents whose average length (2,105.3 words) is four times longer than other benchmark datasets (463.7 for OntoNotes), and contains examples of difficult coreference problems common in literature. This dataset allows for an evaluation of cross-domain performance for the task of coreference resolution, and analysis into the characteristics of long-distance within-document coreference.
△ Less
Submitted 15 May, 2020; v1 submitted 2 December, 2019;
originally announced December 2019.
-
Learning to Groove with Inverse Sequence Transformations
Authors:
Jon Gillick,
Adam Roberts,
Jesse Engel,
Douglas Eck,
David Bamman
Abstract:
We explore models for translating abstract musical ideas (scores, rhythms) into expressive performances using Seq2Seq and recurrent Variational Information Bottleneck (VIB) models. Though Seq2Seq models usually require painstakingly aligned corpora, we show that it is possible to adapt an approach from the Generative Adversarial Network (GAN) literature (e.g. Pix2Pix (Isola et al., 2017) and Vid2V…
▽ More
We explore models for translating abstract musical ideas (scores, rhythms) into expressive performances using Seq2Seq and recurrent Variational Information Bottleneck (VIB) models. Though Seq2Seq models usually require painstakingly aligned corpora, we show that it is possible to adapt an approach from the Generative Adversarial Network (GAN) literature (e.g. Pix2Pix (Isola et al., 2017) and Vid2Vid (Wang et al. 2018a)) to sequences, creating large volumes of paired data by performing simple transformations and training generative models to plausibly invert these transformations. Music, and drumming in particular, provides a strong test case for this approach because many common transformations (quantization, removing voices) have clear semantics, and models for learning to invert them have real-world applications. Focusing on the case of drum set players, we create and release a new dataset for this purpose, containing over 13 hours of recordings by professional drummers aligned with fine-grained timing and dynamics information. We also explore some of the creative potential of these models, including demonstrating improvements on state-of-the-art methods for Humanization (instantiating a performance from a musical score).
△ Less
Submitted 26 July, 2019; v1 submitted 14 May, 2019;
originally announced May 2019.
-
DeepSeek: Content Based Image Search & Retrieval
Authors:
Tanya Piplani,
David Bamman
Abstract:
Most of the internet today is composed of digital media that includes videos and images. With pixels becoming the currency in which most transactions happen on the internet, it is becoming increasingly important to have a way of browsing through this ocean of information with relative ease. YouTube has 400 hours of video uploaded every minute and many million images are browsed on Instagram, Faceb…
▽ More
Most of the internet today is composed of digital media that includes videos and images. With pixels becoming the currency in which most transactions happen on the internet, it is becoming increasingly important to have a way of browsing through this ocean of information with relative ease. YouTube has 400 hours of video uploaded every minute and many million images are browsed on Instagram, Facebook, etc. Inspired by recent advances in the field of deep learning and success that it has gained on various problems like image captioning and, machine translation , word2vec , skip thoughts, etc, we present DeepSeek a natural language processing based deep learning model that allows users to enter a description of the kind of images that they want to search, and in response the system retrieves all the images that semantically and contextually relate to the query. Two approaches are described in the following sections.
△ Less
Submitted 11 January, 2018; v1 submitted 9 January, 2018;
originally announced January 2018.
-
Annotating Character Relationships in Literary Texts
Authors:
Philip Massey,
Patrick Xia,
David Bamman,
Noah A. Smith
Abstract:
We present a dataset of manually annotated relationships between characters in literary texts, in order to support the training and evaluation of automatic methods for relation type prediction in this domain (Makazhanov et al., 2014; Kokkinakis, 2013) and the broader computational analysis of literary character (Elson et al., 2010; Bamman et al., 2014; Vala et al., 2015; Flekova and Gurevych, 2015…
▽ More
We present a dataset of manually annotated relationships between characters in literary texts, in order to support the training and evaluation of automatic methods for relation type prediction in this domain (Makazhanov et al., 2014; Kokkinakis, 2013) and the broader computational analysis of literary character (Elson et al., 2010; Bamman et al., 2014; Vala et al., 2015; Flekova and Gurevych, 2015). In this work, we solicit annotations from workers on Amazon Mechanical Turk for 109 texts ranging from Homer's _Iliad_ to Joyce's _Ulysses_ on four dimensions of interest: for a given pair of characters, we collect judgments as to the coarse-grained category (professional, social, familial), fine-grained category (friend, lover, parent, rival, employer), and affinity (positive, negative, neutral) that describes their primary relationship in a text. We do not assume that this relationship is static; we also collect judgments as to whether it changes at any point in the course of the text.
△ Less
Submitted 2 December, 2015;
originally announced December 2015.
-
A framework for (under)specifying dependency syntax without overloading annotators
Authors:
Nathan Schneider,
Brendan O'Connor,
Naomi Saphra,
David Bamman,
Manaal Faruqui,
Noah A. Smith,
Chris Dyer,
Jason Baldridge
Abstract:
We introduce a framework for lightweight dependency syntax annotation. Our formalism builds upon the typical representation for unlabeled dependencies, permitting a simple notation and annotation workflow. Moreover, the formalism encourages annotators to underspecify parts of the syntax if doing so would streamline the annotation process. We demonstrate the efficacy of this annotation on three lan…
▽ More
We introduce a framework for lightweight dependency syntax annotation. Our formalism builds upon the typical representation for unlabeled dependencies, permitting a simple notation and annotation workflow. Moreover, the formalism encourages annotators to underspecify parts of the syntax if doing so would streamline the annotation process. We demonstrate the efficacy of this annotation on three languages and develop algorithms to evaluate and compare underspecified annotations.
△ Less
Submitted 14 June, 2013; v1 submitted 9 June, 2013;
originally announced June 2013.
-
New Alignment Methods for Discriminative Book Summarization
Authors:
David Bamman,
Noah A. Smith
Abstract:
We consider the unsupervised alignment of the full text of a book with a human-written summary. This presents challenges not seen in other text alignment problems, including a disparity in length and, consequent to this, a violation of the expectation that individual words and phrases should align, since large passages and chapters can be distilled into a single summary phrase. We present two new…
▽ More
We consider the unsupervised alignment of the full text of a book with a human-written summary. This presents challenges not seen in other text alignment problems, including a disparity in length and, consequent to this, a violation of the expectation that individual words and phrases should align, since large passages and chapters can be distilled into a single summary phrase. We present two new methods, based on hidden Markov models, specifically targeted to this problem, and demonstrate gains on an extractive book summarization task. While there is still much room for improvement, unsupervised alignment holds intrinsic value in offering insight into what features of a book are deemed worthy of summarization.
△ Less
Submitted 6 May, 2013;
originally announced May 2013.
-
Inferring Social Rank in an Old Assyrian Trade Network
Authors:
David Bamman,
Adam Anderson,
Noah A. Smith
Abstract:
We present work in jointly inferring the unique individuals as well as their social rank within a collection of letters from an Old Assyrian trade colony in Kültepe, Turkey, settled by merchants from the ancient city of Assur for approximately 200 years between 1950-1750 BCE, the height of the Middle Bronze Age. Using a probabilistic latent-variable model, we leverage pairwise social differences b…
▽ More
We present work in jointly inferring the unique individuals as well as their social rank within a collection of letters from an Old Assyrian trade colony in Kültepe, Turkey, settled by merchants from the ancient city of Assur for approximately 200 years between 1950-1750 BCE, the height of the Middle Bronze Age. Using a probabilistic latent-variable model, we leverage pairwise social differences between names in cuneiform tablets to infer a single underlying social order that best explains the data we observe. Evaluating our output with published judgments by domain experts suggests that our method may be used for building informed hypotheses that are driven by data, and that may offer promising avenues for directed research by Assyriologists.
△ Less
Submitted 12 March, 2013;
originally announced March 2013.
-
Gender identity and lexical variation in social media
Authors:
David Bamman,
Jacob Eisenstein,
Tyler Schnoebelen
Abstract:
We present a study of the relationship between gender, linguistic style, and social networks, using a novel corpus of 14,000 Twitter users. Prior quantitative work on gender often treats this social variable as a female/male binary; we argue for a more nuanced approach. By clustering Twitter users, we find a natural decomposition of the dataset into various styles and topical interests. Many clust…
▽ More
We present a study of the relationship between gender, linguistic style, and social networks, using a novel corpus of 14,000 Twitter users. Prior quantitative work on gender often treats this social variable as a female/male binary; we argue for a more nuanced approach. By clustering Twitter users, we find a natural decomposition of the dataset into various styles and topical interests. Many clusters have strong gender orientations, but their use of linguistic resources sometimes directly conflicts with the population-level language statistics. We view these clusters as a more accurate reflection of the multifaceted nature of gendered language styles. Previous corpus-based work has also had little to say about individuals whose linguistic styles defy population-level gender patterns. To identify such individuals, we train a statistical classifier, and measure the classifier confidence for each individual in the dataset. Examining individuals whose language does not match the classifier's model for their gender, we find that they have social networks that include significantly fewer same-gender social connections and that, in general, social network homophily is correlated with the use of same-gender language markers. Pairing computational methods and social theory thus offers a new perspective on how gender emerges as individuals position themselves relative to audiences, topics, and mainstream gender norms.
△ Less
Submitted 12 May, 2014; v1 submitted 16 October, 2012;
originally announced October 2012.
