Provocations from the Humanities for Generative AI Research

This paper presents a set of provocations for considering the uses, impact, and harms of generative AI from the perspective of humanities researchers. We provide a working definition of humanities research, summarize some of its most salient theories and methods, and apply these theories and methods to the current landscape of AI. Drawing from foundational work in critical data studies, along with relevant humanities scholarship, we elaborate eight claims with broad applicability to current conversations about generative AI: 1) Models make words, but people make meaning; 2) Generative AI requires an expanded definition of culture; 3) Generative AI can never be representative; 4) Bigger models are not always better models; 5) Not all training data is equivalent; 6) Openness is not an easy fix; 7) Limited access to compute enables corporate capture; and 8) AI universalism creates narrow human subjects. We conclude with a discussion of the importance of resisting the extraction of humanities research by computer science and related fields.

Lost in Space: Optimizing Tokens for Grammar-Constrained Decoding

General-purpose language models are trained to produce varied natural language outputs, but for some tasks like annotation or classification we need more specific output formats. LLM systems increasingly support structured output, sampling tokens according to a grammar, which enforces a format but which can also reduce performance. We ask whether there are systematic differences between grammars that appear semantically similar to humans. To answer this question, we test four popular model families with five token formats on four NLP benchmarks. All models perform most accurately when instructed to classify with real numbers. Performance also improves by 5%-10% when models are instructed to return tokens incorporating leading whitespace, which we find can help models avoid structural deficiencies in subword token representations. Format-based differences are largest for smaller models that are often used for local laptop-scale inference. We present best practices for researchers using language models as zero-shot classifiers with structured output.

Looking for the Inner Music: Probing LLMs' Understanding of Literary Style

Recent work has demonstrated that language models can be trained to identify the author of much shorter literary passages than has been thought feasible for traditional stylometry. We replicate these results for authorship and extend them to a new dataset measuring novel genre. We find that LLMs are able to distinguish authorship and genre, but they do so in different ways. Some models seem to rely more on memorization, while others benefit more from training to learn author/genre characteristics. We then use three methods to probe one high-performing LLM for features that define style. These include direct syntactic ablations to input text as well as two methods that look at model internals. We find that authorial style is easier to define than genre-level style and is more impacted by minor syntactic decisions and contextual word usage. However, some traits like pronoun usage and word order prove significant for defining both kinds of literary style.

Automate or Assist? The Role of Computational Models in Identifying Gendered Discourse in US Capital Trial Transcripts

The language used by US courtroom actors in criminal trials has long been studied for biases. However, systematic studies for bias in high-stakes court trials have been difficult, due to the nuanced nature of bias and the legal expertise required. New large language models offer the possibility to automate annotation, saving time and cost. But validating these approaches requires both high quantitative performance as well as an understanding of how automated methods fit in existing workflows, and what they really offer. In this paper we present a case study of adding an automated system to a complex and high-stakes problem: identifying gender-biased language in US capital trials for women defendants. Our team of experienced death-penalty lawyers and NLP technologists pursued a three-phase study: first annotating manually, then training and evaluating computational models, and finally comparing human annotations to model predictions. Unlike many typical NLP tasks, annotating for gender bias in months-long capital trials was a complicated task that involves with many individual judgment calls. In contrast to standard arguments for automation that are based on efficiency and scalability, legal experts found the computational models most useful in challenging their personal bias in annotation and providing opportunities to refine and build consensus on rules for annotation. This suggests that seeking to replace experts with computational models is both unrealistic and undesirable. Rather, computational models offer valuable opportunities to assist the legal experts in annotation-based studies.

How Chinese are Chinese Language Models? The Puzzling Lack of Language Policy in China's LLMs

Contemporary language models are increasingly multilingual, but Chinese LLM developers must navigate complex political and business considerations of language diversity. Language policy in China aims at influencing the public discourse and governing a multi-ethnic society, and has gradually transitioned from a pluralist to a more assimilationist approach since 1949. We explore the impact of these influences on current language technology. We evaluate six open-source multilingual LLMs pre-trained by Chinese companies on 18 languages, spanning a wide range of Chinese, Asian, and Anglo-European languages. Our experiments show Chinese LLMs performance on diverse languages is indistinguishable from international LLMs. Similarly, the models' technical reports also show lack of consideration for pretraining data language coverage except for English and Mandarin Chinese. Examining Chinese AI policy, model experiments, and technical reports, we find no sign of any consistent policy, either for or against, language diversity in China's LLM development. This leaves a puzzling fact that while China regulates both the languages people use daily as well as language model development, they do not seem to have any policy on the languages in language models.

Stronger Random Baselines for In-Context Learning

Evaluating the in-context learning classification performance of language models poses challenges due to small dataset sizes, extensive prompt-selection using the validation set, and intentionally difficult tasks that lead to near-random performance. The standard random baseline -- the expected accuracy of guessing labels uniformly at random -- is stable when the evaluation set is used only once or when the dataset is large. We account for the common practice of validation set reuse and existing small datasets with a stronger random baseline: the expected maximum accuracy across multiple random classifiers. When choosing the best prompt demonstrations across six quantized language models applied to 16 BIG-bench Lite tasks, more than 20\% of the few-shot results that exceed the standard baseline do not exceed this stronger random baseline. When held-out test sets are available, this stronger baseline is also a better predictor of held-out performance than the standard baseline, avoiding unnecessary test set evaluations. This maximum random baseline provides an easily calculated drop-in replacement for the standard baseline.

and and , Oh My! Literary Coreference Annotation with LLMs

Coreference annotation and resolution is a vital component of computational literary studies. However, it has previously been difficult to build high quality systems for fiction. Coreference requires complicated structured outputs, and literary text involves subtle inferences and highly varied language. New language-model-based seq2seq systems present the opportunity to solve both these problems by learning to directly generate a copy of an input sentence with markdown-like annotations. We create, evaluate, and release several trained models for coreference, as well as a workflow for training new models.

The Afterlives of Shakespeare and Company in Online Social Readership

The growth of social reading platforms such as Goodreads and LibraryThing enables us to analyze reading activity at very large scale and in remarkable detail. But twenty-first century systems give us a perspective only on contemporary readers. Meanwhile, the digitization of the lending library records of Shakespeare and Company provides a window into the reading activity of an earlier, smaller community in interwar Paris. In this article, we explore the extent to which we can make comparisons between the Shakespeare and Company and Goodreads communities. By quantifying similarities and differences, we can identify patterns in how works have risen or fallen in popularity across these datasets. We can also measure differences in how works are received by measuring similarities and differences in co-reading patterns. Finally, by examining the complete networks of co-readership, we can observe changes in the overall structures of literary reception.

Hyperpolyglot LLMs: Cross-Lingual Interpretability in Token Embeddings

Cross-lingual transfer learning is an important property of multilingual large language models (LLMs). But how do LLMs represent relationships between languages? Every language model has an input layer that maps tokens to vectors. This ubiquitous layer of language models is often overlooked. We find that similarities between these input embeddings are highly interpretable and that the geometry of these embeddings differs between model families. In one case (XLM-RoBERTa), embeddings encode language: tokens in different writing systems can be linearly separated with an average of 99.2% accuracy. Another family (mT5) represents cross-lingual semantic similarity: the 50 nearest neighbors for any token represent an average of 7.61 writing systems, and are frequently translations. This result is surprising given that there is no explicit parallel cross-lingual training corpora and no explicit incentive for translations in pre-training objectives. Our research opens the door for investigations in 1) The effect of pre-training and model architectures on representations of languages and 2) The applications of cross-lingual representations embedded in language models.

Data Similarity is Not Enough to Explain Language Model Performance

Large language models achieve high performance on many but not all downstream tasks. The interaction between pretraining data and task data is commonly assumed to determine this variance: a task with data that is more similar to a model's pretraining data is assumed to be easier for that model. We test whether distributional and example-specific similarity measures (embedding-, token- and model-based) correlate with language model performance through a large-scale comparison of the Pile and C4 pretraining datasets with downstream benchmarks. Similarity correlates with performance for multilingual datasets, but in other benchmarks, we surprisingly find that similarity metrics are not correlated with accuracy or even each other. This suggests that the relationship between pretraining data and downstream tasks is more complex than often assumed.