Computation and Language (cs.CL)

Abstract: Large Language Models have shown impressive abilities on various tasks. However, fundamentally improving them depends on high-quality datasets or computationally expensive fine-tuning. On the contrary, human can easily improve themselves by thinking and memory, without external resources. In this paper, we propose a framework, MoT, to let the LLM self-improve through Memory of Thoughts, without annotated datasets and parameter updates. Specifically, the framework is divided into two stages: 1. before the test stage, we let the LLM pre-think on the unlabeled dataset and save the high-confidence thoughts as external memory; 2. during inference, given a test question, we let the LLM recall relevant memory to help itself reason and answer it. Experimental results show that the proposed framework can help ChatGPT significantly improve its abilities in math reasoning, commonsense reasoning, factual reasoning and natural language inference. Further analyses show that each component contributes critically to the improvements.

Abstract: Recently, much Chinese text error correction work has focused on Chinese Spelling Check (CSC) and Chinese Grammatical Error Diagnosis (CGED). In contrast, little attention has been paid to the complicated problem of Chinese Semantic Error Diagnosis (CSED), which lacks relevant datasets. The study of semantic errors is important because they are very common and may lead to syntactic irregularities or even problems of comprehension. To investigate this, we build the CSED corpus, which includes two datasets. The one is for the CSED-Recognition (CSED-R) task. The other is for the CSED-Correction (CSED-C) task. Our annotation guarantees high-quality data through quality assurance mechanisms. Our experiments show that powerful pre-trained models perform poorly on this corpus. We also find that the CSED task is challenging, as evidenced by the fact that even humans receive a low score. This paper proposes syntax-aware models to specifically adapt to the CSED task. The experimental results show that the introduction of the syntax-aware approach is meaningful.

Abstract: Diffusion models, which have emerged to become popular text-to-image generation models, can produce high-quality and content-rich images guided by textual prompts. However, there are limitations to semantic understanding and commonsense reasoning in existing models when the input prompts are concise narrative, resulting in low-quality image generation. To improve the capacities for narrative prompts, we propose a simple-yet-effective parameter-efficient fine-tuning approach called the Semantic Understanding and Reasoning adapter (SUR-adapter) for pre-trained diffusion models. To reach this goal, we first collect and annotate a new dataset SURD which consists of more than 57,000 semantically corrected multi-modal samples. Each sample contains a simple narrative prompt, a complex keyword-based prompt, and a high-quality image. Then, we align the semantic representation of narrative prompts to the complex prompts and transfer knowledge of large language models (LLMs) to our SUR-adapter via knowledge distillation so that it can acquire the powerful semantic understanding and reasoning capabilities to build a high-quality textual semantic representation for text-to-image generation. We conduct experiments by integrating multiple LLMs and popular pre-trained diffusion models to show the effectiveness of our approach in enabling diffusion models to understand and reason concise natural language without image quality degradation. Our approach can make text-to-image diffusion models easier to use with better user experience, which demonstrates our approach has the potential for further advancing the development of user-friendly text-to-image generation models by bridging the semantic gap between simple narrative prompts and complex keyword-based prompts.

Abstract: Detecting commonsense causal relations (causation) between events has long been an essential yet challenging task. Given that events are complicated, an event may have different causes under various contexts. Thus, exploiting context plays an essential role in detecting causal relations. Meanwhile, previous works about commonsense causation only consider two events and ignore their context, simplifying the task formulation. This paper proposes a new task to detect commonsense causation between two events in an event sequence (i.e., context), called contextualized commonsense causal reasoning. We also design a zero-shot framework: COLA (Contextualized Commonsense Causality Reasoner) to solve the task from the causal inference perspective. This framework obtains rich incidental supervision from temporality and balances covariates from multiple timestamps to remove confounding effects. Our extensive experiments show that COLA can detect commonsense causality more accurately than baselines.

Abstract: Self-supervised learning (SSL) has been dramatically successful not only in monolingual but also in cross-lingual settings. However, since the two settings have been studied individually in general, there has been little research focusing on how effective a cross-lingual model is in comparison with a monolingual model. In this paper, we investigate this fundamental question empirically with Japanese automatic speech recognition (ASR) tasks. First, we begin by comparing the ASR performance of cross-lingual and monolingual models for two different language tasks while keeping the acoustic domain as identical as possible. Then, we examine how much unlabeled data collected in Japanese is needed to achieve performance comparable to a cross-lingual model pre-trained with tens of thousands of hours of English and/or multilingual data. Finally, we extensively investigate the effectiveness of SSL in Japanese and demonstrate state-of-the-art performance on multiple ASR tasks. Since there is no comprehensive SSL study for Japanese, we hope this study will guide Japanese SSL research.

Abstract: We address the task of machine translation from an extremely low-resource language (LRL) to English using cross-lingual transfer from a closely related high-resource language (HRL). For many of these languages, no parallel corpora are available, even monolingual corpora are limited and representations in pre-trained sequence-to-sequence models are absent. These factors limit the benefits of cross-lingual transfer from shared embedding spaces in multilingual models. However, many extremely LRLs have a high level of lexical similarity with related HRLs. We utilize this property by injecting character and character-span noise into the training data of the HRL prior to learning the vocabulary. This serves as a regularizer which makes the model more robust to lexical divergences between the HRL and LRL and better facilitates cross-lingual transfer. On closely related HRL and LRL pairs from multiple language families, we observe that our method significantly outperforms the baseline MT as well as approaches proposed previously to address cross-lingual transfer between closely related languages. We also show that the proposed character-span noise injection performs better than the unigram-character noise injection.

Abstract: Text image machine translation (TIMT) has been widely used in various real-world applications, which translates source language texts in images into another target language sentence. Existing methods on TIMT are mainly divided into two categories: the recognition-then-translation pipeline model and the end-to-end model. However, how to transfer knowledge from the pipeline model into the end-to-end model remains an unsolved problem. In this paper, we propose a novel Multi-Teacher Knowledge Distillation (MTKD) method to effectively distillate knowledge into the end-to-end TIMT model from the pipeline model. Specifically, three teachers are utilized to improve the performance of the end-to-end TIMT model. The image encoder in the end-to-end TIMT model is optimized with the knowledge distillation guidance from the recognition teacher encoder, while the sequential encoder and decoder are improved by transferring knowledge from the translation sequential and decoder teacher models. Furthermore, both token and sentence-level knowledge distillations are incorporated to better boost the translation performance. Extensive experimental results show that our proposed MTKD effectively improves the text image translation performance and outperforms existing end-to-end and pipeline models with fewer parameters and less decoding time, illustrating that MTKD can take advantage of both pipeline and end-to-end models.

Abstract: In everyday life, humans often plan their actions by following step-by-step instructions in the form of goal-oriented scripts. Previous work has exploited language models (LMs) to plan for abstract goals of stereotypical activities (e.g., "make a cake"), but leaves more specific goals with multi-facet constraints understudied (e.g., "make a cake for diabetics"). In this paper, we define the task of constrained language planning for the first time. We propose an overgenerate-then-filter approach to improve large language models (LLMs) on this task, and use it to distill a novel constrained language planning dataset, CoScript, which consists of 55,000 scripts. Empirical results demonstrate that our method significantly improves the constrained language planning ability of LLMs, especially on constraint faithfulness. Furthermore, CoScript is demonstrated to be quite effective in endowing smaller LMs with constrained language planning ability.

Abstract: Named Entity Recognition (NER) is a cornerstone NLP task while its robustness has been given little attention. This paper rethinks the principles of NER attacks derived from sentence classification, as they can easily violate the label consistency between the original and adversarial NER examples. This is due to the fine-grained nature of NER, as even minor word changes in the sentence can result in the emergence or mutation of any entities, resulting in invalid adversarial examples. To this end, we propose a novel one-word modification NER attack based on a key insight, NER models are always vulnerable to the boundary position of an entity to make their decision. We thus strategically insert a new boundary into the sentence and trigger the Entity Boundary Interference that the victim model makes the wrong prediction either on this boundary word or on other words in the sentence. We call this attack Virtual Boundary Attack (ViBA), which is shown to be remarkably effective when attacking both English and Chinese models with a 70%-90% attack success rate on state-of-the-art language models (e.g. RoBERTa, DeBERTa) and also significantly faster than previous methods.

Abstract: Attention-based contextual biasing approaches have shown significant improvements in the recognition of generic and/or personal rare-words in End-to-End Automatic Speech Recognition (E2E ASR) systems like neural transducers. These approaches employ cross-attention to bias the model towards specific contextual entities injected as bias-phrases to the model. Prior approaches typically relied on subword encoders for encoding the bias phrases. However, subword tokenizations are coarse and fail to capture granular pronunciation information which is crucial for biasing based on acoustic similarity. In this work, we propose to use lightweight character representations to encode fine-grained pronunciation features to improve contextual biasing guided by acoustic similarity between the audio and the contextual entities (termed acoustic biasing). We further integrate pretrained neural language model (NLM) based encoders to encode the utterance's semantic context along with contextual entities to perform biasing informed by the utterance's semantic context (termed semantic biasing). Experiments using a Conformer Transducer model on the Librispeech dataset show a 4.62% - 9.26% relative WER improvement on different biasing list sizes over the baseline contextual model when incorporating our proposed acoustic and semantic biasing approach. On a large-scale in-house dataset, we observe 7.91% relative WER improvement compared to our baseline model. On tail utterances, the improvements are even more pronounced with 36.80% and 23.40% relative WER improvements on Librispeech rare words and an in-house testset respectively.

Abstract: Compared to news and chat summarization, the development of meeting summarization is hugely decelerated by the limited data. To this end, we introduce a versatile Chinese meeting summarization dataset, dubbed VCSum, consisting of 239 real-life meetings, with a total duration of over 230 hours. We claim our dataset is versatile because we provide the annotations of topic segmentation, headlines, segmentation summaries, overall meeting summaries, and salient sentences for each meeting transcript. As such, the dataset can adapt to various summarization tasks or methods, including segmentation-based summarization, multi-granularity summarization and retrieval-then-generate summarization. Our analysis confirms the effectiveness and robustness of VCSum. We also provide a set of benchmark models regarding different downstream summarization tasks on VCSum to facilitate further research. The dataset and code will be released at \url{https://github.com/hahahawu/VCSum}.

Abstract: Goal-directed dialogue systems aim to proactively reach a pre-determined target through multi-turn conversations. The key to achieving this task lies in planning dialogue paths that smoothly and coherently direct conversations towards the target. However, this is a challenging and under-explored task. In this work, we propose a coherent dialogue planning approach that uses a stochastic process to model the temporal dynamics of dialogue paths. We define a latent space that captures the coherence of goal-directed behavior using a Brownian bridge process, which allows us to incorporate user feedback flexibly in dialogue planning. Based on the derived latent trajectories, we generate dialogue paths explicitly using pre-trained language models. We finally employ these paths as natural language prompts to guide dialogue generation. Our experiments show that our approach generates more coherent utterances and achieves the goal with a higher success rate.

Abstract: Transferring information retrieval (IR) models from a high-resource language (typically English) to other languages in a zero-shot fashion has become a widely adopted approach. In this work, we show that the effectiveness of zero-shot rankers diminishes when queries and documents are present in different languages. Motivated by this, we propose to train ranking models on artificially code-switched data instead, which we generate by utilizing bilingual lexicons. To this end, we experiment with lexicons induced from (1) cross-lingual word embeddings and (2) parallel Wikipedia page titles. We use the mMARCO dataset to extensively evaluate reranking models on 36 language pairs spanning Monolingual IR (MoIR), Cross-lingual IR (CLIR), and Multilingual IR (MLIR). Our results show that code-switching can yield consistent and substantial gains of 5.1 MRR@10 in CLIR and 3.9 MRR@10 in MLIR, while maintaining stable performance in MoIR. Encouragingly, the gains are especially pronounced for distant languages (up to 2x absolute gain). We further show that our approach is robust towards the ratio of code-switched tokens and also extends to unseen languages. Our results demonstrate that training on code-switched data is a cheap and effective way of generalizing zero-shot rankers for cross-lingual and multilingual retrieval.

Abstract: We develop computational models to analyze court statements in order to assess judicial attitudes toward victims of sexual violence in the Israeli court system. The study examines the resonance of "rape myths" in the criminal justice system's response to sex crimes, in particular in judicial assessment of victim's credibility. We begin by formulating an ontology for evaluating judicial attitudes toward victim's credibility, with eight ordinal labels and binary categorizations. Second, we curate a manually annotated dataset for judicial assessments of victim's credibility in the Hebrew language, as well as a model that can extract credibility labels from court cases. The dataset consists of 855 verdict decision documents in sexual assault cases from 1990-2021, annotated with the help of legal experts and trained law students. The model uses a combined approach of syntactic and latent structures to find sentences that convey the judge's attitude towards the victim and classify them according to the credibility label set. Our ontology, data, and models will be made available upon request, in the hope they spur future progress in this judicial important task.

Abstract: Structured sentiment analysis (SSA) aims to automatically extract people's opinions from a text in natural language and adequately represent that information in a graph structure. One of the most accurate methods for performing SSA was recently proposed and consists of approaching it as a dependency parsing task. Although we can find in the literature how transition-based algorithms excel in dependency parsing in terms of accuracy and efficiency, all proposed attempts to tackle SSA following that approach were based on graph-based models. In this article, we present the first transition-based method to address SSA as dependency parsing. Specifically, we design a transition system that processes the input text in a left-to-right pass, incrementally generating the graph structure containing all identified opinions. To effectively implement our final transition-based model, we resort to a Pointer Network architecture as a backbone. From an extensive evaluation, we demonstrate that our model offers the best performance to date in practically all cases among prior dependency-based methods, and surpass recent task-specific techniques on the most challenging datasets. We additionally include an in-depth analysis and empirically prove that the overall time-complexity cost of our approach is quadratic in the sentence length, being more efficient than top-performing graph-based parsers.

Abstract: As the impact of technology on our lives is increasing, we witness increased use of social media that became an essential tool not only for communication but also for sharing information with community about our thoughts and feelings. This can be observed also for people with mental health disorders such as depression where they use social media for expressing their thoughts and asking for help. This opens a possibility to automatically process social media posts and detect signs of depression. We build several large pre-trained language model based classifiers for depression detection from social media posts. Besides fine-tuning BERT, RoBERTA, BERTweet, and mentalBERT were also construct two types of ensembles. We analyze the performance of our models on two data sets of posts from social platforms Reddit and Twitter, and investigate also the performance of transfer learning across the two data sets. The results show that transformer ensembles improve over the single transformer-based classifiers.

Abstract: Effective argumentation is essential towards a purposeful conversation with a satisfactory outcome. For example, persuading someone to reconsider smoking might involve empathetic, well founded arguments based on facts and expert opinions about its ill-effects and the consequences on one's family. However, the automatic generation of high-quality factual arguments can be challenging. Addressing existing controllability issues can make the recent advances in computational models for argument generation a potential solution. In this paper, we introduce ArgU: a neural argument generator capable of producing factual arguments from input facts and real-world concepts that can be explicitly controlled for stance and argument structure using Walton's argument scheme-based control codes. Unfortunately, computational argument generation is a relatively new field and lacks datasets conducive to training. Hence, we have compiled and released an annotated corpora of 69,428 arguments spanning six topics and six argument schemes, making it the largest publicly available corpus for identifying argument schemes; the paper details our annotation and dataset creation framework. We further experiment with an argument generation strategy that establishes an inference strategy by generating an ``argument template'' before actual argument generation. Our results demonstrate that it is possible to automatically generate diverse arguments exhibiting different inference patterns for the same set of facts by using control codes based on argument schemes and stance.

Abstract: The subtle human values we acquire through life experiences govern our thoughts and gets reflected in our speech. It plays an integral part in capturing the essence of our individuality and making it imperative to identify such values in computational systems that mimic human actions. Computational argumentation is a field that deals with the argumentation capabilities of humans and can benefit from identifying such values. Motivated by that, we present an ensemble approach for detecting human values from argument text. Our ensemble comprises three models: (i) An entailment-based model for determining the human values based on their descriptions, (ii) A Roberta-based classifier that predicts the set of human values from an argument. (iii) A Roberta-based classifier to predict a reduced set of human values from an argument. We experiment with different ways of combining the models and report our results. Furthermore, our best combination achieves an overall F1 score of 0.48 on the main test set.

Abstract: The number of web pages is growing at an exponential rate, accumulating massive amounts of data on the web. It is one of the key processes to classify webpages in web information mining. Some classical methods are based on manually building features of web pages and training classifiers based on machine learning or deep learning. However, building features manually requires specific domain knowledge and usually takes a long time to validate the validity of features. Considering webpages generated by the combination of text and HTML Document Object Model(DOM) trees, we propose a representation and classification method based on a pre-trained language model and graph neural network, named PLM-GNN. It is based on the joint encoding of text and HTML DOM trees in the web pages. It performs well on the KI-04 and SWDE datasets and on practical dataset AHS for the project of scholar's homepage crawling.

Abstract: Theory of mind (ToM) refers to humans' ability to understand and infer the desires, beliefs, and intentions of others. The acquisition of ToM plays a key role in humans' social cognition and interpersonal relations. Though indispensable for social intelligence, ToM is still lacking for modern AI and NLP systems since they cannot access the human mental state and cognitive process beneath the training corpus. To empower AI systems with the ToM ability and narrow the gap between them and humans, in this paper, we propose COKE: the first cognitive knowledge graph for machine theory of mind. Specifically, COKE formalizes ToM as a collection of 45k+ manually verified cognitive chains that characterize human mental activities and subsequent behavioral/affective responses when facing specific social circumstances. Beyond that, we further generalize COKE using pre-trained language models and build a powerful cognitive generation model COKE+. Experimental results in both automatic and human evaluation demonstrate the high quality of COKE and the superior ToM ability of COKE+.

Abstract: The medical conversational question answering (CQA) system aims at providing a series of professional medical services to improve the efficiency of medical care. Despite the success of large language models (LLMs) in complex reasoning tasks in various fields, such as mathematics, logic, and commonsense QA, they still need to improve with the increased complexity and specialization of the medical field. This is because medical CQA tasks require not only strong medical reasoning, but also the ability to think broadly and deeply. In this paper, to address these challenges in medical CQA tasks that need to be considered and understood in many aspects, we propose the Holistically Thought (HoT) method, which is designed to guide the LLMs to perform the diffused and focused thinking for generating high-quality medical responses. The proposed HoT method has been evaluated through automated and manual assessments in three different medical CQA datasets containing the English and Chinese languages. The extensive experimental results show that our method can produce more correctness, professional, and considerate answers than several state-of-the-art (SOTA) methods, manifesting its effectiveness. Our code in https://github.com/WENGSYX/HoT.

Abstract: 'Mahabharata' is the most popular among many Indian pieces of literature referred to in many domains for completely different purposes. This text itself is having various dimension and aspects which is useful for the human being in their personal life and professional life. This Indian Epic is originally written in the Sanskrit Language. Now in the era of Natural Language Processing, Artificial Intelligence, Machine Learning, and Human-Computer interaction this text can be processed according to the domain requirement. It is interesting to process this text and get useful insights from Mahabharata. The limitation of the humans while analyzing Mahabharata is that they always have a sentiment aspect towards the story narrated by the author. Apart from that, the human cannot memorize statistical or computational details, like which two words are frequently coming in one sentence? What is the average length of the sentences across the whole literature? Which word is the most popular word across the text, what are the lemmas of the words used across the sentences? Thus, in this paper, we propose an NLP pipeline to get some statistical and computational insights along with the most relevant word searching method from the largest epic 'Mahabharata'. We stacked the different text-processing approaches to articulate the best results which can be further used in the various domain where Mahabharata needs to be referred.

Abstract: Webpages have been a rich resource for language and vision-language tasks. Yet only pieces of webpages are kept: image-caption pairs, long text articles, or raw HTML, never all in one place. Webpage tasks have resultingly received little attention and structured image-text data underused. To study multimodal webpage understanding, we introduce the Wikipedia Webpage 2M (WikiWeb2M) suite; the first to retain the full set of images, text, and structure data available in a page. WikiWeb2M can be used for tasks like page description generation, section summarization, and contextual image captioning.

Abstract: This paper aims to benchmark recent progress in language understanding models that output contextualised representations at the character level. Many such modelling architectures and methods to train those architectures have been proposed, but it is currently unclear what the relative contributions of the architecture vs. the pretraining objective are to final model performance. We explore the design space of such models, comparing architectural innovations and a variety of different pretraining objectives on a suite of evaluation tasks with a fixed training procedure in order to find the currently optimal way to build and train character-level BERT-like models. We find that our best performing character-level model exceeds the performance of a token-based model trained with the same settings on the same data, suggesting that character-level models are ready for more widespread adoption. Unfortunately, the best method to train character-level models still relies on a subword-level tokeniser during pretraining, and final model performance is highly dependent on tokeniser quality. We believe our results demonstrate the readiness of character-level models for multilingual language representation, and encourage NLP practitioners to try them as drop-in replacements for token-based models.

Abstract: With the recent advances in natural language processing (NLP), a vast number of applications have emerged across various use cases. Among the plethora of NLP applications, many academic researchers are motivated to do work that has a positive social impact, in line with the recent initiatives of NLP for Social Good (NLP4SG). However, it is not always obvious to researchers how their research efforts are tackling today's big social problems. Thus, in this paper, we introduce NLP4SGPAPERS, a scientific dataset with three associated tasks that can help identify NLP4SG papers and characterize the NLP4SG landscape by: (1) identifying the papers that address a social problem, (2) mapping them to the corresponding UN Sustainable Development Goals (SDGs), and (3) identifying the task they are solving and the methods they are using. Using state-of-the-art NLP models, we address each of these tasks and use them on the entire ACL Anthology, resulting in a visualization workspace that gives researchers a comprehensive overview of the field of NLP4SG. Our website is available at https://nlp4sg.vercel.app . We released our data at https://huggingface.co/datasets/feradauto/NLP4SGPapers and code at https://github.com/feradauto/nlp4sg .

Abstract: Novel intent discovery automates the process of grouping similar messages (questions) to identify previously unknown intents. However, current research focuses on publicly available datasets which have only the question field and significantly differ from real-life datasets. This paper proposes methods to improve the intent discovery pipeline deployed in a large e-commerce platform. We show the benefit of pre-training language models on in-domain data: both self-supervised and with weak supervision. We also devise the best method to utilize the conversational structure (i.e., question and answer) of real-life datasets during fine-tuning for clustering tasks, which we call Conv. All our methods combined to fully utilize real-life datasets give up to 33pp performance boost over state-of-the-art Constrained Deep Adaptive Clustering (CDAC) model for question only. By comparison CDAC model for the question data only gives only up to 13pp performance boost over the naive baseline.

Abstract: We would like to explore how morphemes can affect the performance of a language model. We trained GPT-2 and Bert model with StateMorph for both Finnish and Russian, which is a morpheme segmenting algorithm. As a comparison, we also trained a model with BPE and Morfessor. Our preliminary result shows that StateMorph can help the model to converge more efficiently and achieve a better validation score.

Abstract: Recently, open-domain question answering systems have begun to rely heavily on annotated datasets to train neural passage retrievers. However, manually annotating such datasets is both difficult and time-consuming, which limits their availability for less popular languages. In this work, we experiment with several methods for automatically collecting weakly labeled datasets and show how they affect the performance of the neural passage retrieval models. As a result of our work, we publish the MAUPQA dataset, consisting of nearly 400,000 question-passage pairs for Polish, as well as the HerBERT-QA neural retriever.

Abstract: Cross-modal contrastive learning in vision language pretraining (VLP) faces the challenge of (partial) false negatives. In this paper, we study this problem from the perspective of Mutual Information (MI) optimization. It is common sense that InfoNCE loss used in contrastive learning will maximize the lower bound of MI between anchors and their positives, while we theoretically prove that MI involving negatives also matters when noises commonly exist. Guided by a more general lower bound form for optimization, we propose a contrastive learning strategy regulated by progressively refined cross-modal similarity, to more accurately optimize MI between an image/text anchor and its negative texts/images instead of improperly minimizing it. Our method performs competitively on four downstream cross-modal tasks and systematically balances the beneficial and harmful effects of (partial) false negative samples under theoretical guidance.

Abstract: Large Language Models (LLMs) have made rapid progress in recent months and weeks, garnering significant public attention. This has sparked concerns about aligning these models with human values, their impact on labor markets, and the potential need for regulation in further research and development. However, the discourse often lacks a focus on the imperative to widely diffuse the societal benefits of LLMs. To qualify this societal benefit, we assert that LLMs exhibit emergent abilities to humanize technology more effectively than previous technologies, and for people across language, occupation, and accessibility divides. We argue that they do so by addressing three mechanizing bottlenecks in today's computing technologies: creating diverse and accessible content, learning complex digital tools, and personalizing machine learning algorithms. We adopt a case-based approach and illustrate each bottleneck with two examples where current technology imposes bottlenecks that LLMs demonstrate the ability to address. Given this opportunity to humanize technology widely, we advocate for more widespread understanding of LLMs, tools and methods to simplify use of LLMs, and cross-cutting institutional capacity.

Abstract: This work explores the utility of explicit structure for representation learning in NLP by developing StrAE -- an autoencoding framework that faithfully leverages sentence structure to learn multi-level node embeddings in an unsupervised fashion. We use StrAE to train models across different types of sentential structure and objectives, including a novel contrastive loss over structure, and evaluate the learnt embeddings on a series of both intrinsic and extrinsic tasks. Our experiments indicate that leveraging explicit structure through StrAE leads to improved embeddings over prior work, and that our novel contrastive objective over structure outperforms the standard cross-entropy objective. Moreover, in contrast to findings from prior work that weakly leverages structure, we find that being completely faithful to structure does enable disambiguation between types of structure based on the corresponding model's performance. As further evidence of StrAE's utility, we develop a simple proof-of-concept approach to simultaneously induce structure while learning embeddings, rather than being given structure, and find that performance is comparable to that of the best-performing models where structure is given. Finally, we contextualise these results by comparing StrAE against standard unstructured baselines learnt in similar settings, and show that faithfully leveraging explicit structure can be beneficial in lexical and sentence-level semantics.

Abstract: Background: Artificial intelligence language models have shown promise in various applications, including assisting with clinical decision-making as demonstrated by strong performance of large language models on medical licensure exams. However, their ability to solve complex, open-ended cases, which may be representative of clinical practice, remains unexplored. Methods: In this study, the accuracy of large language AI models GPT4 and GPT3.5 in diagnosing complex clinical cases was investigated using published Case Records of the Massachusetts General Hospital. A total of 50 cases requiring a diagnosis and diagnostic test published from January 1, 2022 to April 16, 2022 were identified. For each case, models were given a prompt requesting the top three specific diagnoses and associated diagnostic tests, followed by case text, labs, and figure legends. Model outputs were assessed in comparison to the final clinical diagnosis and whether the model-predicted test would result in a correct diagnosis. Results: GPT4 and GPT3.5 accurately provided the correct diagnosis in 26% and 22% of cases in one attempt, and 46% and 42% within three attempts, respectively. GPT4 and GPT3.5 provided a correct essential diagnostic test in 28% and 24% of cases in one attempt, and 44% and 50% within three attempts, respectively. No significant differences were found between the two models, and multiple trials with identical prompts using the GPT3.5 model provided similar results. Conclusions: In summary, these models demonstrate potential usefulness in generating differential diagnoses but remain limited in their ability to provide a single unifying diagnosis in complex, open-ended cases. Future research should focus on evaluating model performance in larger datasets of open-ended clinical challenges and exploring potential human-AI collaboration strategies to enhance clinical decision-making.

Abstract: Standard methods for multi-label text classification largely rely on encoder-only pre-trained language models, whereas encoder-decoder models have proven more effective in other classification tasks. In this study, we compare four methods for multi-label classification, two based on an encoder only, and two based on an encoder-decoder. We carry out experiments on four datasets -- two in the legal domain and two in the biomedical domain, each with two levels of label granularity -- and always depart from the same pre-trained model, T5. Our results show that encoder-decoder methods outperform encoder-only methods, with a growing advantage on more complex datasets and labeling schemes of finer granularity. Using encoder-decoder models in a non-autoregressive fashion, in particular, yields the best performance overall, so we further study this approach through ablations to better understand its strengths.

Abstract: While ``instruction-tuned" generative large language models (LLMs) have demonstrated an impressive ability to generalize to new tasks, the training phases heavily rely on large amounts of diverse and high-quality instruction data (such as ChatGPT and GPT-4). Unfortunately, acquiring high-quality data, especially when it comes to human-written data, can pose significant challenges both in terms of cost and accessibility. Moreover, concerns related to privacy can further limit access to such data, making the process of obtaining it a complex and nuanced undertaking. Consequently, this hinders the generality of the tuned models and may restrict their effectiveness in certain contexts. To tackle this issue, our study introduces a new approach called Federated Instruction Tuning (FedIT), which leverages federated learning (FL) as the learning framework for the instruction tuning of LLMs. This marks the first exploration of FL-based instruction tuning for LLMs. This is especially important since text data is predominantly generated by end users. Therefore, it is imperative to design and adapt FL approaches to effectively leverage these users' diverse instructions stored on local devices, while preserving privacy and ensuring data security. In the current paper, by conducting widely used GPT-4 auto-evaluation, we demonstrate that by exploiting the heterogeneous and diverse sets of instructions on the client's end with the proposed framework FedIT, we improved the performance of LLMs compared to centralized training with only limited local instructions. Further, in this paper, we developed a Github repository named Shepherd. This repository offers a foundational framework for exploring federated fine-tuning of LLMs using heterogeneous instructions across diverse categories.