Recent work using Rank-One Model Editing (ROME), a popular model editing method, has shown that there are certain facts that the algorithm is unable to edit without breaking the model. Such edits have previously been called disabling edits. These disabling edits cause immediate model collapse and limits the use of ROME for sequential editing. In this paper, we show that disabling edits are an artifact of irregularities in the implementation of ROME. With this paper, we provide a more stable implementation ROME, which we call r-ROME and show that model collapse is no longer observed when making large scale sequential edits with r-ROME, while further improving generalization and locality of model editing compared to the original implementation of ROME. We also provide a detailed mathematical explanation of the reason behind disabling edits.

Editing knowledge in large language models is an attractive capability that allows us to correct incorrectly learned facts during pre-training, as well as update the model with an ever-growing list of new facts. While existing model editing techniques have shown promise, they are usually evaluated using metrics for reliability, specificity and generalization over one or few edits. We argue that for model editing to have practical utility, we must be able to make multiple edits to the same model. With this in mind, we evaluate current model editing methods at scale, focusing on two state of the art methods - ROME and MEMIT. With the lens of scalability, we evaluate model editing methods for three crucial properties - editing proficiency, fact forgetting and downstream performance. We find that as a model is edited sequentially with multiple facts, it continually becomes less editable, forgets previously edited facts and loses the ability to perform downstream tasks. For ROME and MEMIT, this “forgetting” happens in two phases - an initial gradual but progressive forgetting phase followed by an abrupt or catastrophic forgetting. Both gradual and catastrophic forgetting limit the usefulness of model editing methods at scale - the former makes model editing less effective as multiple edits are made to the model while the latter caps the scalability of such model editing methods. Our analysis also highlights other key limitations of ROME and MEMIT at scale. With our work, we push for better evaluation of model editing and development of model editing methods keeping scalability in mind.

ROME and MEMIT are largely believed to be two different model editing algorithms, with the major difference between them being the ability to perform batched edits. In this paper, we unify these two algorithms under a single conceptual umbrella, optimizing for the same goal, which we call the preservation-memorization objective. ROME uses an equality constraint to optimize this objective to perform one edit at a time, whereas MEMIT employs a more flexible least-square constraint that allows for batched edits. We generalize ROME and enable batched editing with equality constraint in the form of EMMET - an Equality-constrained Mass Model Editing algorithm for Transformers, a new batched memory-editing algorithm. EMMET can perform batched-edits up to a batch-size of 10,000, with very similar performance to MEMIT across multiple dimensions. With the introduction of EMMET, we truly unify ROME and MEMIT and show that both algorithms are equivalent in terms of their optimization objective, their abilities (singular and batched editing), their model editing performance and their limitations.

As large language models (LLM) evolve in their capabilities, various recent studies have tried to quantify their behavior using psychological tools created to study human behavior. One such example is the measurement of “personality” of LLMs using self-assessment personality tests developed to measure human personality. Yet almost none of these works verify the applicability of these tests on LLMs. In this paper, we analyze the reliability of LLM personality scores obtained from self-assessment personality tests using two simple experiments. We first introduce the property of prompt sensitivity, where three semantically equivalent prompts representing three intuitive ways of administering self-assessment tests on LLMs are used to measure the personality of the same LLM. We find that all three prompts lead to very different personality scores, a difference that is statistically significant for all traits in a large majority of scenarios. We then introduce the property of option-order symmetry for personality measurement of LLMs. Since most of the self-assessment tests exist in the form of multiple choice question (MCQ) questions, we argue that the scores should also be robust to not just the prompt template but also the order in which the options are presented. This test unsurprisingly reveals that the self-assessment test scores are not robust to the order of the options. These simple tests, done on ChatGPT and three Llama2 models of different sizes, show that self-assessment personality tests created for humans are unreliable measures of personality in LLMs.

Domain adaptation is an important and widely studied problem in natural language processing. A large body of literature tries to solve this problem by adapting models trained on the source domain to the target domain. In this paper, we instead solve this problem from a dataset perspective. We modify the source domain dataset with simple lexical transformations to reduce the domain shift between the source dataset distribution and the target dataset distribution. We find that models trained on the transformed source domain dataset performs significantly better than zero-shot models. Using our proposed transformations to convert standard English to tweets, we reach an unsupervised part-of-speech (POS) tagging accuracy of 92.14% (from 81.54% zero shot accuracy), which is only slightly below the supervised performance of 94.45%. We also use our proposed transformations to synthetically generate tweets and augment the Twitter dataset to achieve state-of-the-art performance for POS tagging.

With their increasing size, large language models (LLMs) are becoming increasingly good at language understanding tasks. But even with high performance on specific downstream task, LLMs fail at simple linguistic tests for negation or quantifier understanding. Previous work on quantifier understanding in LLMs show inverse scaling in understanding few-type quantifiers. In this paper, we question the claims of of previous work and show that it is a result of inappropriate testing methodology. We also present alternate methods to measure quantifier comprehension in LLMs and show that LLMs are able to better understand the difference between the meaning of few-type and most-type quantifiers as their size increases, although they are not particularly good at it. We also observe inverse scaling for most-type quantifier understanding, which is contrary to human psycho-linguistic experiments and previous work, where the model’s understanding of most-type quantifier gets worse as the model size increases. We do this evaluation on models ranging from 125M-175B parameters, which suggests that LLMs do not do as well as expected with quantifiers. We also discuss the possible reasons for this and the relevance of quantifier understanding in evaluating language understanding in LLMs.

Stock sentiment has strong correlations with the stock market but traditional sentiment analysis task classifies sentiment according to having feelings and emotions of good or bad. This definition of sentiment is not an accurate indicator of public opinion about specific stocks. To bridge this gap, we introduce a new task of stock sentiment analysis and present a new dataset for this task named TweetFinSent. In TweetFinSent, tweets are annotated based on if one gained or expected to gain positive or negative return from a stock. Experiments on TweetFinSent with several sentiment analysis models from lexicon-based to transformer-based have been conducted. Experimental results show that TweetFinSent dataset constitutes a challenging problem and there is ample room for improvement on the stock sentiment analysis task. TweetFinSent is available at https://github.com/jpmcair/tweetfinsent.

Spoken dialog systems are slowly becoming an integral part of the human experience due to their various advantages over textual interfaces. Spoken language understanding (SLU) systems are fundamental building blocks of spoken dialog systems. But creating SLU systems for low resourced languages is still a challenge. In a large number of low resourced language, we don’t have access to enough data to build automatic speech recognition (ASR) technologies, which are fundamental to any SLU system. Also, ASR based SLU systems do not generalize to unwritten languages. In this paper, we present a series of experiments to explore extremely low-resourced settings where we perform intent classification with systems trained on as low as one data-point per intent and with only one speaker in the dataset. We also work in a low-resourced setting where we do not use language specific ASR systems to transcribe input speech, which compounds the challenge of building SLU systems to simulate a true low-resourced setting. We test our system on Belgian Dutch (Flemish) and English and find that using phonetic transcriptions to make intent classification systems in such low-resourced setting performs significantly better than using speech features. Specifically, when using a phonetic transcription based system over a feature based system, we see average improvements of 12.37% and 13.08% for binary and four-class classification problems respectively, when averaged over 49 different experimental settings.

In hope of expanding training data, researchers often want to merge two or more datasets that are created using different labeling schemes. This paper considers two datasets that label part-of-speech (POS) tags under different tagging schemes and leverage the supervised labels of one dataset to help generate labels for the other dataset. This paper further discusses the theoretical difficulties of this approach and proposes a novel supervised architecture employing Transformers to tackle the problem of consolidating two completely disjoint datasets. The results diverge from initial expectations and discourage exploration into the use of disjoint labels to consolidate datasets with different labels.

This paper presents our submission for the SMM4H 2022-Shared Task on the classification of self-reported intimate partner violence on Twitter (in English). The goal of this task was to accurately determine if the contents of a given tweet demonstrated someone reporting their own experience with intimate partner violence. The submitted system is an ensemble of five RoBERTa models each weighted by their respective F1-scores on the validation data-set. This system performed 13% better than the baseline and was the best performing system overall for this shared task.

We present our response to Task 5 of the Social Media Mining for Health Applications (SMM4H) 2022 competition. We share our approach into classifying whether a tweet in Spanish about COVID-19 symptoms pertain to themselves, others, or not at all. Using a combination of BERT based models, we were able to achieve results that were higher than the median result of the competition.

This paper presents my submission for Tasks 1 and 2 for the Social Media Mining of Health (SMM4H) 2022 Shared Tasks competition. I first describe the background behind each of these tasks, followed by the descriptions of the various subtasks of Tasks 1 and 2, then present the methodology. Through model ensembling, this methodology was able to achieve higher results than the mean and median of the competition for the classification tasks.

Sentiment analysis is an important task in understanding social media content like customer reviews, Twitter and Facebook feeds etc. In multilingual communities around the world, a large amount of social media text is characterized by the presence of Code-Switching. Thus, it has become important to build models that can handle code-switched data. However, annotated code-switched data is scarce and there is a need for unsupervised models and algorithms. We propose a general framework called Unsupervised Self-Training and show its applications for the specific use case of sentiment analysis of code-switched data. We use the power of pre-trained BERT models for initialization and fine-tune them in an unsupervised manner, only using pseudo labels produced by zero-shot transfer. We test our algorithm on multiple code-switched languages and provide a detailed analysis of the learning dynamics of the algorithm with the aim of answering the question - ‘Does our unsupervised model understand the Code-Switched languages or does it just learn its representations?’. Our unsupervised models compete well with their supervised counterparts, with their performance reaching within 1-7% (weighted F1 scores) when compared to supervised models trained for a two class problem.

Sentiment analysis in Code-Mixed languages has garnered a lot of attention in recent years. It is an important task for social media monitoring and has many applications, as a large chunk of social media data is Code-Mixed. In this paper, we work on the problem of sentiment analysis for Dravidian Code-Switched languages - Tamil-Engish and Malayalam-English, using three different BERT based models. We leverage task-specific pre-training and cross-lingual transfer to improve on previously reported results, with significant improvement for the Tamil-Engish dataset. We also present a multilingual sentiment classification model that has competitive performance on both Tamil-English and Malayalam-English datasets.

In this paper we present our submission for the EACL 2021-Shared Task on Offensive Language Identification in Dravidian languages. Our final system is an ensemble of mBERT and XLM-RoBERTa models which leverage task-adaptive pre-training of multilingual BERT models with a masked language modeling objective. Our system was ranked 1st for Kannada, 2nd for Malayalam and 3rd for Tamil.