Robust, faithful and harm-free pronoun use for individuals is an important goal for language models as their use increases, but prior work tends to study only one or two of these components at a time. To measure progress towards the combined goal, we introduce the task of pronoun use fidelity: given a context introducing a co-referring entity and pronoun, the task is to reuse the correct pronoun later, independent of potential distractors. We present a carefully-designed dataset of over 5 million instances to evaluate pronoun use fidelity in English, and we use it to evaluate 37 popular large language models across architectures (encoder-only, decoder-only and encoder-decoder) and scales (11M-70B parameters). We find that while models can mostly faithfully reuse previously-specified pronouns in the presence of no distractors, they are significantly worse at processing she/her/her, singular they and neopronouns. Additionally, models are not robustly faithful to pronouns, as they are easily distracted. With even one additional sentence containing a distractor pronoun, accuracy drops on average by 34%. With 5 distractor sentences, accuracy drops by 52% for decoder-only models and 13% for encoder-only models. We show that widely-used large language models are still brittle, with large gaps in reasoning and in processing different pronouns in a setting that is very simple for humans, and we encourage researchers in bias and reasoning to bridge them.
ORCA (Shen et al., 2023) is a recent technique for cross-modal fine-tuning, i.e., applying pre-trained transformer models to modalities beyond their training data. The technique consists primarily of training an embedder and fine-tuning the embedder and model. Despite its high performance on a variety of downstream tasks, we do not understand precisely how each of these components contribute to ORCA's success. Therefore, we run a series of ablations and find that embedder training does not help 2D tasks at all, contrary to what the original paper posits. In 1D tasks, some amount of embedder training is necessary but more is not better. In 4 out of 6 datasets we experiment with, it is model fine-tuning that makes the biggest difference. Through our ablations and baselines, we contribute a better understanding of the individual components of ORCA.
In-context learning is a popular inference strategy where large language models solve a task using only a few labelled demonstrations without needing any parameter updates. Compared to work on monolingual (English) in-context learning, multilingual in-context learning is under-explored, and we lack an in-depth understanding of the role of demonstrations in this context. To address this gap, we conduct a multidimensional analysis of multilingual in-context learning, experimenting with 5 models from different model families, 9 datasets covering classification and generation tasks, and 56 typologically diverse languages. Our results reveal that the effectiveness of demonstrations varies significantly across models, tasks, and languages. We also find that Llama 2-Chat, GPT-3.5, and GPT-4 are largely insensitive to the quality of demonstrations. Instead, a carefully crafted template often eliminates the benefits of demonstrations for some tasks and languages altogether. These findings show that the importance of demonstrations might be overestimated. Our work highlights the need for granular evaluation across multiple axes towards a better understanding of in-context learning.
If a question cannot be answered with the available information, robust systems for question answering (QA) should know _not_ to answer. One way to build QA models that do this is with additional training data comprised of unanswerable questions, created either by employing annotators or through automated methods for unanswerable question generation. To show that the model complexity of existing automated approaches is not justified, we examine a simpler data augmentation method for unanswerable question generation in English: performing antonym and entity swaps on answerable questions. Compared to the prior state-of-the-art, data generated with our training-free and lightweight strategy results in better models (+1.6 F1 points on SQuAD 2.0 data with BERT-large), and has higher human-judged relatedness and readability. We quantify the raw benefits of our approach compared to no augmentation across multiple encoder models, using different amounts of generated data, and also on TydiQA-MinSpan data (+9.3 F1 points with BERT-large). Our results establish swaps as a simple but strong baseline for future work.
Intersectionality is a critical framework that, through inquiry and praxis, allows us to examine how social inequalities persist through domains of structure and discipline. Given AI fairness' raison d'\^etre of ``fairness,'' we argue that adopting intersectionality as an analytical framework is pivotal to effectively operationalizing fairness. Through a critical review of how intersectionality is discussed in 30 papers from the AI fairness literature, we deductively and inductively: 1) map how intersectionality tenets operate within the AI fairness paradigm and 2) uncover gaps between the conceptualization and operationalization of intersectionality. We find that researchers overwhelmingly reduce intersectionality to optimizing for fairness metrics over demographic subgroups. They also fail to discuss their social context and when mentioning power, they mostly situate it only within the AI pipeline. We: 3) outline and assess the implications of these gaps for critical inquiry and praxis, and 4) provide actionable recommendations for AI fairness researchers to engage with intersectionality in their work by grounding it in AI epistemology.