Computation and Language (cs.CL)

Abstract: This paper presents a modified neural model for topic detection from a corpus and proposes a new metric to evaluate the detected topics. The new model builds upon the embedded topic model incorporating some modifications such as document clustering. Numerical experiments suggest that the new model performs favourably regardless of the document's length. The new metric, which can be computed more efficiently than widely-used metrics such as topic coherence, provides variable information regarding the understandability of the detected topics.

Abstract: The existing supervised relation extraction methods have achieved impressive performance in a closed-set setting, where the relations during both training and testing remain the same. In a more realistic open-set setting, unknown relations may appear in the test set. Due to the lack of supervision signals from unknown relations, a well-performing closed-set relation extractor can still confidently misclassify them into known relations. In this paper, we propose an unknown-aware training method, regularizing the model by dynamically synthesizing negative instances. To facilitate a compact decision boundary, ``difficult'' negative instances are necessary. Inspired by text adversarial attacks, we adaptively apply small but critical perturbations to original training instances and thus synthesizing negative instances that are more likely to be mistaken by the model as known relations. Experimental results show that this method achieves SOTA unknown relation detection without compromising the classification of known relations.

Abstract: Semantic matching is a mainstream paradigm of zero-shot relation extraction, which matches a given input with a corresponding label description. The entities in the input should exactly match their hypernyms in the description, while the irrelevant contexts should be ignored when matching. However, general matching methods lack explicit modeling of the above matching pattern. In this work, we propose a fine-grained semantic matching method tailored for zero-shot relation extraction. Following the above matching pattern, we decompose the sentence-level similarity score into entity and context matching scores. Due to the lack of explicit annotations of the redundant components, we design a feature distillation module to adaptively identify the relation-irrelevant features and reduce their negative impact on context matching. Experimental results show that our method achieves higher matching $F_1$ score and has an inference speed 10 times faster, when compared with the state-of-the-art methods.

Abstract: The usage of more than one language in the same text is referred to as Code Mixed. It is evident that there is a growing degree of adaption of the use of code-mixed data, especially English with a regional language, on social media platforms. Existing deep-learning models do not take advantage of the implicit language information in the code-mixed text. Our study aims to improve BERT-based models performance on low-resource Code-Mixed Hindi-English Datasets by experimenting with language augmentation approaches. We propose a pipeline to improve code-mixed systems that comprise data preprocessing, word-level language identification, language augmentation, and model training on downstream tasks like sentiment analysis. For language augmentation in BERT models, we explore word-level interleaving and post-sentence placement of language information. We have examined the performance of vanilla BERT-based models and their code-mixed HingBERT counterparts on respective benchmark datasets, comparing their results with and without using word-level language information. The models were evaluated using metrics such as accuracy, precision, recall, and F1 score. Our findings show that the proposed language augmentation approaches work well across different BERT models. We demonstrate the importance of augmenting code-mixed text with language information on five different code-mixed Hindi-English downstream datasets based on sentiment analysis, hate speech detection, and emotion detection.

Abstract: Current clustering-based Open Relation Extraction (OpenRE) methods usually adopt a two-stage pipeline. The first stage simultaneously learns relation representations and assignments. The second stage manually labels several instances and thus names the relation for each cluster. However, unsupervised objectives struggle to optimize the model to derive accurate clustering assignments, and the number of clusters has to be supplied in advance. In this paper, we present a novel setting, named actively supervised clustering for OpenRE. Our insight lies in that clustering learning and relation labeling can be alternately performed, providing the necessary guidance for clustering without a significant increase in human effort. The key to the setting is selecting which instances to label. Instead of using classical active labeling strategies designed for fixed known classes, we propose a new strategy, which is applicable to dynamically discover clusters of unknown relations. Experimental results show that our method is able to discover almost all relational clusters in the data and improve the SOTA methods by 10.3\% and 5.2\%, on two datasets respectively.

Abstract: This work introduces approaches to assessing phrase breaks in ESL learners' speech using pre-trained language models (PLMs) and large language models (LLMs). There are two tasks: overall assessment of phrase break for a speech clip and fine-grained assessment of every possible phrase break position. To leverage NLP models, speech input is first force-aligned with texts, and then pre-processed into a token sequence, including words and phrase break information. To utilize PLMs, we propose a pre-training and fine-tuning pipeline with the processed tokens. This process includes pre-training with a replaced break token detection module and fine-tuning with text classification and sequence labeling. To employ LLMs, we design prompts for ChatGPT. The experiments show that with the PLMs, the dependence on labeled training data has been greatly reduced, and the performance has improved. Meanwhile, we verify that ChatGPT, a renowned LLM, has potential for further advancement in this area.

Abstract: Cross-lingual text classification leverages text classifiers trained in a high-resource language to perform text classification in other languages with no or minimal fine-tuning (zero/few-shots cross-lingual transfer). Nowadays, cross-lingual text classifiers are typically built on large-scale, multilingual language models (LMs) pretrained on a variety of languages of interest. However, the performance of these models vary significantly across languages and classification tasks, suggesting that the superposition of the language modelling and classification tasks is not always effective. For this reason, in this paper we propose revisiting the classic "translate-and-test" pipeline to neatly separate the translation and classification stages. The proposed approach couples 1) a neural machine translator translating from the targeted language to a high-resource language, with 2) a text classifier trained in the high-resource language, but the neural machine translator generates "soft" translations to permit end-to-end backpropagation during fine-tuning of the pipeline. Extensive experiments have been carried out over three cross-lingual text classification datasets (XNLI, MLDoc and MultiEURLEX), with the results showing that the proposed approach has significantly improved performance over a competitive baseline.

Abstract: Large language models (LLMs)have achieved great success in general domains of natural language processing. In this paper, we bring LLMs to the realm of geoscience, with the objective of advancing research and applications in this field. To this end, we present the first-ever LLM in geoscience, K2, alongside a suite of resources developed to further promote LLM research within geoscience. For instance, we have curated the first geoscience instruction tuning dataset, GeoSignal, which aims to align LLM responses to geoscience-related user queries. Additionally, we have established the first geoscience benchmark, GeoBenchmark, to evaluate LLMs in the context of geoscience. In this work, we experiment with a complete recipe to adapt a pretrained general-domain LLM to the geoscience domain. Specifically, we further train the LLaMA-7B model on over 1 million pieces of geoscience literature and utilize GeoSignal's supervised data to fine-tune the model. Moreover, we share a protocol that can efficiently gather domain-specific data and construct domain-supervised data, even in situations where manpower is scarce. Experiments conducted on the GeoBenchmark demonstrate the the effectiveness of our approach and datasets.

Abstract: Misogyny and sexism are growing problems in social media. Advances have been made in online sexism detection but the systems are often uninterpretable. SemEval-2023 Task 10 on Explainable Detection of Online Sexism aims at increasing explainability of the sexism detection, and our team participated in all the proposed subtasks. Our system is based on further domain-adaptive pre-training (Gururangan et al., 2020). Building on the Transformer-based models with the domain adaptation, we compare fine-tuning with multi-task learning and show that each subtask requires a different system configuration. In our experiments, multi-task learning performs on par with standard fine-tuning for sexism detection and noticeably better for coarse-grained sexism classification, while fine-tuning is preferable for fine-grained classification.

Abstract: A few benchmarking datasets have been released to evaluate the factual knowledge of pretrained language models. These benchmarks (e.g., LAMA, and ParaRel) are mainly developed in English and later are translated to form new multilingual versions (e.g., mLAMA, and mParaRel). Results on these multilingual benchmarks suggest that using English prompts to recall the facts from multilingual models usually yields significantly better and more consistent performance than using non-English prompts. Our analysis shows that mLAMA is biased toward facts from Western countries, which might affect the fairness of probing models. We propose a new framework for curating factual triples from Wikidata that are culturally diverse. A new benchmark DLAMA-v1 is built of factual triples from three pairs of contrasting cultures having a total of 78,259 triples from 20 relation predicates. The three pairs comprise facts representing the (Arab and Western), (Asian and Western), and (South American and Western) countries respectively. Having a more balanced benchmark (DLAMA-v1) supports that mBERT performs better on Western facts than non-Western ones, while monolingual Arabic, English, and Korean models tend to perform better on their culturally proximate facts. Moreover, both monolingual and multilingual models tend to make a prediction that is culturally or geographically relevant to the correct label, even if the prediction is wrong.

Abstract: The integration of language models for neural machine translation has been extensively studied in the past. It has been shown that an external language model, trained on additional target-side monolingual data, can help improve translation quality. However, there has always been the assumption that the translation model also learns an implicit target-side language model during training, which interferes with the external language model at decoding time. Recently, some works on automatic speech recognition have demonstrated that, if the implicit language model is neutralized in decoding, further improvements can be gained when integrating an external language model. In this work, we transfer this concept to the task of machine translation and compare with the most prominent way of including additional monolingual data - namely back-translation. We find that accounting for the implicit language model significantly boosts the performance of language model fusion, although this approach is still outperformed by back-translation.

Abstract: Existing sentence textual similarity benchmark datasets only use a single number to summarize how similar the sentence encoder's decision is to humans'. However, it is unclear what kind of sentence pairs a sentence encoder (SE) would consider similar. Moreover, existing SE benchmarks mainly consider sentence pairs with low lexical overlap, so it is unclear how the SEs behave when two sentences have high lexical overlap. We introduce a high-quality SE diagnostic dataset, HEROS. HEROS is constructed by transforming an original sentence into a new sentence based on certain rules to form a \textit{minimal pair}, and the minimal pair has high lexical overlaps. The rules include replacing a word with a synonym, an antonym, a typo, a random word, and converting the original sentence into its negation. Different rules yield different subsets of HEROS. By systematically comparing the performance of over 60 supervised and unsupervised SEs on HEROS, we reveal that most unsupervised sentence encoders are insensitive to negation. We find the datasets used to train the SE are the main determinants of what kind of sentence pairs an SE considers similar. We also show that even if two SEs have similar performance on STS benchmarks, they can have very different behavior on HEROS. Our result reveals the blind spot of traditional STS benchmarks when evaluating SEs.

Abstract: Instruction tuning large language models (LLMs) remains a challenging task, owing to the complexity of hyperparameter selection and the difficulty involved in evaluating the tuned models. To determine the optimal hyperparameters, an automatic, robust, and reliable evaluation benchmark is essential. However, establishing such a benchmark is not a trivial task due to the challenges associated with evaluation accuracy and privacy protection. In response to these challenges, we introduce a judge large language model, named PandaLM, which is trained to distinguish the superior model given several LLMs. PandaLM's focus extends beyond just the objective correctness of responses, which is the main focus of traditional evaluation datasets. It addresses vital subjective factors such as relative conciseness, clarity, adherence to instructions, comprehensiveness, and formality. To ensure the reliability of PandaLM, we collect a diverse human-annotated test dataset, where all contexts are generated by humans and labels are aligned with human preferences. Our results indicate that PandaLM-7B achieves 93.75% of GPT-3.5's evaluation ability and 88.28% of GPT-4's in terms of F1-score on our test dataset. PandaLM enables the evaluation of LLM to be fairer but with less cost, evidenced by significant improvements achieved by models tuned through PandaLM compared to their counterparts trained with default Alpaca's hyperparameters. In addition, PandaLM does not depend on API-based evaluations, thus avoiding potential data leakage. All resources of PandaLM are released at https://github.com/WeOpenML/PandaLM.

Abstract: Phonetic convergence describes the automatic and unconscious speech adaptation of two interlocutors in a conversation. This paper proposes a Siamese recurrent neural network (RNN) architecture to measure the convergence of the holistic spectral characteristics of speech sounds in an L2-L2 interaction. We extend an alternating reading task (the ART) dataset by adding 20 native Slovak L2 English speakers. We train and test the Siamese RNN model to measure phonetic convergence of L2 English speech from three different native language groups: Italian (9 dyads), French (10 dyads) and Slovak (10 dyads). Our results indicate that the Siamese RNN model effectively captures the dynamics of phonetic convergence and the speaker's imitation ability. Moreover, this text-independent model is scalable and capable of handling L1-induced speaker variability.

Abstract: Regulatory bodies worldwide are intensifying their efforts to ensure transparency in influencer marketing on social media through instruments like the Unfair Commercial Practices Directive (UCPD) in the European Union, or Section 5 of the Federal Trade Commission Act. Yet enforcing these obligations has proven to be highly problematic due to the sheer scale of the influencer market. The task of automatically detecting sponsored content aims to enable the monitoring and enforcement of such regulations at scale. Current research in this field primarily frames this problem as a machine learning task, focusing on developing models that achieve high classification performance in detecting ads. These machine learning tasks rely on human data annotation to provide ground truth information. However, agreement between annotators is often low, leading to inconsistent labels that hinder the reliability of models. To improve annotation accuracy and, thus, the detection of sponsored content, we propose using chatGPT to augment the annotation process with phrases identified as relevant features and brief explanations. Our experiments show that this approach consistently improves inter-annotator agreement and annotation accuracy. Additionally, our survey of user experience in the annotation task indicates that the explanations improve the annotators' confidence and streamline the process. Our proposed methods can ultimately lead to more transparency and alignment with regulatory requirements in sponsored content detection.

Abstract: Compared to sentence-level systems, document-level neural machine translation (NMT) models produce a more consistent output across a document and are able to better resolve ambiguities within the input. There are many works on document-level NMT, mostly focusing on modifying the model architecture or training strategy to better accommodate the additional context-input. On the other hand, in most works, the question on how to perform search with the trained model is scarcely discussed, sometimes not mentioned at all. In this work, we aim to answer the question how to best utilize a context-aware translation model in decoding. We start with the most popular document-level NMT approach and compare different decoding schemes, some from the literature and others proposed by us. In the comparison, we are using both, standard automatic metrics, as well as specific linguistic phenomena on three standard document-level translation benchmarks. We find that most commonly used decoding strategies perform similar to each other and that higher quality context information has the potential to further improve the translation.

Abstract: Factuality is important to dialogue summarization. Factual error correction (FEC) of model-generated summaries is one way to improve factuality. Current FEC evaluation that relies on factuality metrics is not reliable and detailed enough. To address this problem, we are the first to manually annotate a FEC dataset for dialogue summarization containing 4000 items and propose FERRANTI, a fine-grained evaluation framework based on reference correction that automatically evaluates the performance of FEC models on different error categories. Using this evaluation framework, we conduct sufficient experiments with FEC approaches under a variety of settings and find the best training modes and significant differences in the performance of the existing approaches on different factual error categories.

Abstract: Over the years, many researchers have seemingly made the same observation: Brain and language model activations exhibit some structural similarities, enabling linear partial mappings between features extracted from neural recordings and computational language models. In an attempt to evaluate how much evidence has been accumulated for this observation, we survey over 30 studies spanning 10 datasets and 8 metrics. How much evidence has been accumulated, and what, if anything, is missing before we can draw conclusions? Our analysis of the evaluation methods used in the literature reveals that some of the metrics are less conservative. We also find that the accumulated evidence, for now, remains ambiguous, but correlations with model size and quality provide grounds for cautious optimism.

Abstract: Owing to the impressive dot-product attention, the Transformers have been the dominant architectures in various natural language processing (NLP) tasks. Recently, the Receptance Weighted Key Value (RWKV) architecture follows a non-transformer architecture to eliminate the drawbacks of dot-product attention, where memory and computational complexity exhibits quadratic scaling with sequence length. Although RWKV has exploited a linearly tensor-product attention mechanism and achieved parallelized computations by deploying the time-sequential mode, it fails to capture long-range dependencies because of its limitation on looking back at previous information, compared with full information obtained by direct interactions in the standard transformer. Therefore, the paper devises the Retrospected Receptance Weighted Key Value (RRWKV) architecture via incorporating the retrospecting ability into the RWKV to effectively absorb information, which maintains memory and computational efficiency as well.

Abstract: Despite the existence of various benchmarks for evaluating natural language processing models, we argue that human exams are a more suitable means of evaluating general intelligence for large language models (LLMs), as they inherently demand a much wider range of abilities such as language understanding, domain knowledge, and problem-solving skills. To this end, we introduce M3Exam, a novel benchmark sourced from real and official human exam questions for evaluating LLMs in a multilingual, multimodal, and multilevel context. M3Exam exhibits three unique characteristics: (1) multilingualism, encompassing questions from multiple countries that require strong multilingual proficiency and cultural knowledge; (2) multimodality, accounting for the multimodal nature of many exam questions to test the model's multimodal understanding capability; and (3) multilevel structure, featuring exams from three critical educational periods to comprehensively assess a model's proficiency at different levels. In total, M3Exam contains 12,317 questions in 9 diverse languages with three educational levels, where about 23\% of the questions require processing images for successful solving. We assess the performance of top-performing LLMs on M3Exam and find that current models, including GPT-4, still struggle with multilingual text, particularly in low-resource and non-Latin script languages. Multimodal LLMs also perform poorly with complex multimodal questions. We believe that M3Exam can be a valuable resource for comprehensively evaluating LLMs by examining their multilingual and multimodal abilities and tracking their development. Data and evaluation code is available at \url{https://github.com/DAMO-NLP-SG/M3Exam}.

Abstract: Document-level context for neural machine translation (NMT) is crucial to improve the translation consistency and cohesion, the translation of ambiguous inputs, as well as several other linguistic phenomena. Many works have been published on the topic of document-level NMT, but most restrict the system to only local context, typically including just the one or two preceding sentences as additional information. This might be enough to resolve some ambiguous inputs, but it is probably not sufficient to capture some document-level information like the topic or style of a conversation. When increasing the context size beyond just the local context, there are two challenges: (i) the~memory usage increases exponentially (ii) the translation performance starts to degrade. We argue that the widely-used attention mechanism is responsible for both issues. Therefore, we propose a constrained attention variant that focuses the attention on the most relevant parts of the sequence, while simultaneously reducing the memory consumption. For evaluation, we utilize targeted test sets in combination with novel evaluation techniques to analyze the translations in regards to specific discourse-related phenomena. We find that our approach is a good compromise between sentence-level NMT vs attending to the full context, especially in low resource scenarios.

Abstract: Intelligent systems that aim at mastering language as humans do must deal with its semantic underspecification, namely, the possibility for a linguistic signal to convey only part of the information needed for communication to succeed. Consider the usages of the pronoun they, which can leave the gender and number of its referent(s) underspecified. Semantic underspecification is not a bug but a crucial language feature that boosts its storage and processing efficiency. Indeed, human speakers can quickly and effortlessly integrate semantically-underspecified linguistic signals with a wide range of non-linguistic information, e.g., the multimodal context, social or cultural conventions, and shared knowledge. Standard NLP models have, in principle, no or limited access to such extra information, while multimodal systems grounding language into other modalities, such as vision, are naturally equipped to account for this phenomenon. However, we show that they struggle with it, which could negatively affect their performance and lead to harmful consequences when used for applications. In this position paper, we argue that our community should be aware of semantic underspecification if it aims to develop language technology that can successfully interact with human users. We discuss some applications where mastering it is crucial and outline a few directions toward achieving this goal.

Abstract: The BioNLP Workshop 2023 initiated the launch of a shared task on Problem List Summarization (ProbSum) in January 2023. The aim of this shared task is to attract future research efforts in building NLP models for real-world diagnostic decision support applications, where a system generating relevant and accurate diagnoses will augment the healthcare providers decision-making process and improve the quality of care for patients. The goal for participants is to develop models that generated a list of diagnoses and problems using input from the daily care notes collected from the hospitalization of critically ill patients. Eight teams submitted their final systems to the shared task leaderboard. In this paper, we describe the tasks, datasets, evaluation metrics, and baseline systems. Additionally, the techniques and results of the evaluation of the different approaches tried by the participating teams are summarized.

Abstract: Adverse Event (ADE) extraction is one of the core tasks in digital pharmacovigilance, especially when applied to informal texts. This task has been addressed by the Natural Language Processing community using large pre-trained language models, such as BERT. Despite the great number of Transformer-based architectures used in the literature, it is unclear which of them has better performances and why. Therefore, in this paper we perform an extensive evaluation and analysis of 19 Transformer-based models for ADE extraction on informal texts. We compare the performance of all the considered models on two datasets with increasing levels of informality (forums posts and tweets). We also combine the purely Transformer-based models with two commonly-used additional processing layers (CRF and LSTM), and analyze their effect on the models performance. Furthermore, we use a well-established feature importance technique (SHAP) to correlate the performance of the models with a set of features that describe them: model category (AutoEncoding, AutoRegressive, Text-to-Text), pretraining domain, training from scratch, and model size in number of parameters. At the end of our analyses, we identify a list of take-home messages that can be derived from the experimental data.

Abstract: We consider the task of few-shot intent detection, which involves training a deep learning model to classify utterances based on their underlying intents using only a small amount of labeled data. The current approach to address this problem is through continual pre-training, i.e., fine-tuning pre-trained language models (PLMs) on external resources (e.g., conversational corpora, public intent detection datasets, or natural language understanding datasets) before using them as utterance encoders for training an intent classifier. In this paper, we show that continual pre-training may not be essential, since the overfitting problem of PLMs on this task may not be as serious as expected. Specifically, we find that directly fine-tuning PLMs on only a handful of labeled examples already yields decent results compared to methods that employ continual pre-training, and the performance gap diminishes rapidly as the number of labeled data increases. To maximize the utilization of the limited available data, we propose a context augmentation method and leverage sequential self-distillation to boost performance. Comprehensive experiments on real-world benchmarks show that given only two or more labeled samples per class, direct fine-tuning outperforms many strong baselines that utilize external data sources for continual pre-training. The code can be found at https://github.com/hdzhang-code/DFTPlus.

Abstract: Enabling large language models to effectively utilize real-world tools is crucial for achieving embodied intelligence. Existing approaches to tool learning have primarily relied on either extremely large language models, such as GPT-4, to attain generalized tool-use abilities in a zero-shot manner, or have utilized supervised learning to train limited types of tools on compact models. However, it remains uncertain whether smaller language models can achieve generalized tool-use abilities without specific tool-specific training. To address this question, this paper introduces ToolAlpaca, a novel framework designed to automatically generate a tool-use corpus and learn generalized tool-use abilities on compact language models with minimal human intervention. Specifically, ToolAlpaca first collects a comprehensive dataset by building a multi-agent simulation environment, which contains 3938 tool-use instances from more than 400 real-world tool APIs spanning 50 distinct categories. Subsequently, the constructed corpus is employed to fine-tune compact language models, resulting in two models, namely ToolAlpaca-7B and ToolAlpaca-13B, respectively. Finally, we evaluate the ability of these models to utilize previously unseen tools without specific training. Experimental results demonstrate that ToolAlpaca achieves effective generalized tool-use capabilities comparable to those of extremely large language models like GPT-3.5. This validation supports the notion that learning generalized tool-use abilities is feasible for compact language models.

Abstract: This paper presents novel experiments shedding light on the shortcomings of current metrics for assessing biases of gender discrimination made by machine learning algorithms on textual data. We focus on the Bios dataset, and our learning task is to predict the occupation of individuals, based on their biography. Such prediction tasks are common in commercial Natural Language Processing (NLP) applications such as automatic job recommendations. We address an important limitation of theoretical discussions dealing with group-wise fairness metrics: they focus on large datasets, although the norm in many industrial NLP applications is to use small to reasonably large linguistic datasets for which the main practical constraint is to get a good prediction accuracy. We then question how reliable are different popular measures of bias when the size of the training set is simply sufficient to learn reasonably accurate predictions. Our experiments sample the Bios dataset and learn more than 200 models on different sample sizes. This allows us to statistically study our results and to confirm that common gender bias indices provide diverging and sometimes unreliable results when applied to relatively small training and test samples. This highlights the crucial importance of variance calculations for providing sound results in this field.

Abstract: In this paper, we consider the challenge of summarizing patients' medical progress notes in a limited data setting. For the Problem List Summarization (shared task 1A) at the BioNLP Workshop 2023, we demonstrate that Clinical-T5 fine-tuned to 765 medical clinic notes outperforms other extractive, abstractive and zero-shot baselines, yielding reasonable baseline systems for medical note summarization. Further, we introduce Hierarchical Ensemble of Summarization Models (HESM), consisting of token-level ensembles of diverse fine-tuned Clinical-T5 models, followed by Minimum Bayes Risk (MBR) decoding. Our HESM approach lead to a considerable summarization performance boost, and when evaluated on held-out challenge data achieved a ROUGE-L of 32.77, which was the best-performing system at the top of the shared task leaderboard.

Abstract: Many existing speech translation benchmarks focus on native-English speech in high-quality recording conditions, which often do not match the conditions in real-life use-cases. In this paper, we describe our speech translation system for the multilingual track of IWSLT 2023, which focuses on the translation of scientific conference talks. The test condition features accented input speech and terminology-dense contents. The tasks requires translation into 10 languages of varying amounts of resources. In absence of training data from the target domain, we use a retrieval-based approach (kNN-MT) for effective adaptation (+0.8 BLEU for speech translation). We also use adapters to easily integrate incremental training data from data augmentation, and show that it matches the performance of re-training. We observe that cascaded systems are more easily adaptable towards specific target domains, due to their separate modules. Our cascaded speech system substantially outperforms its end-to-end counterpart on scientific talk translation, although their performance remains similar on TED talks.

Abstract: The introduction of computerized medical records in hospitals has reduced burdensome operations like manual writing and information fetching. However, the data contained in medical records are still far underutilized, primarily because extracting them from unstructured textual medical records takes time and effort. Information Extraction, a subfield of Natural Language Processing, can help clinical practitioners overcome this limitation, using automated text-mining pipelines. In this work, we created the first Italian neuropsychiatric Named Entity Recognition dataset, PsyNIT, and used it to develop a Large Language Model for this task. Moreover, we conducted several experiments with three external independent datasets to implement an effective multicenter model, with overall F1-score 84.77%, Precision 83.16%, Recall 86.44%. The lessons learned are: (i) the crucial role of a consistent annotation process and (ii) a fine-tuning strategy that combines classical methods with a "few-shot" approach. This allowed us to establish methodological guidelines that pave the way for future implementations in this field and allow Italian hospitals to tap into important research opportunities.

Abstract: This paper presents the ADAIO team's system entry in the Building Educational Applications (BEA) 2023 Shared Task on Generating AI Teacher Responses in Educational Dialogues. The task aims to assess the performance of state-of-the-art generative models as AI teachers in producing suitable responses within a student-teacher dialogue. Our system comprises evaluating various baseline models using OpenAI GPT-3 and designing diverse prompts to prompt the OpenAI models for teacher response generation. After the challenge, our system achieved second place by employing a few-shot prompt-based approach with the OpenAI text-davinci-003 model. The results highlight the few-shot learning capabilities of large-language models, particularly OpenAI's GPT-3, in the role of AI teachers.

Abstract: Emerging psychopathology studies are showing that patterns of changes in emotional state -- emotion dynamics -- are associated with overall well-being and mental health. More recently, there has been some work in tracking emotion dynamics through one's utterances, allowing for data to be collected on a larger scale across time and people. However, several questions about how emotion dynamics change with age, especially in children, and when determined through children's writing, remain unanswered. In this work, we use both a lexicon and a machine learning based approach to quantify characteristics of emotion dynamics determined from poems written by children of various ages. We show that both approaches point to similar trends: consistent increasing intensities for some emotions (e.g., anger, fear, joy, sadness, arousal, and dominance) with age and a consistent decreasing valence with age. We also find increasing emotional variability, rise rates (i.e., emotional reactivity), and recovery rates (i.e., emotional regulation) with age. These results act as a useful baselines for further research in how patterns of emotions expressed by children change with age, and their association with mental health.

Abstract: We present a framework that formulates visual question answering as modular code generation. In contrast to prior work on modular approaches to VQA, our approach requires no additional training and relies on pre-trained language models (LMs), visual models pre-trained on image-caption pairs, and fifty VQA examples used for in-context learning. The generated Python programs invoke and compose the outputs of the visual models using arithmetic and conditional logic. Our approach improves accuracy on the COVR dataset by at least 3% and on the GQA dataset by roughly 2% compared to the few-shot baseline that does not employ code generation.

Abstract: Pre-trained language models (PLMs) demonstrate excellent abilities to understand texts in the generic domain while struggling in a specific domain. Although continued pre-training on a large domain-specific corpus is effective, it is costly to tune all the parameters on the domain. In this paper, we investigate whether we can adapt PLMs both effectively and efficiently by only tuning a few parameters. Specifically, we decouple the feed-forward networks (FFNs) of the Transformer architecture into two parts: the original pre-trained FFNs to maintain the old-domain knowledge and our novel domain-specific adapters to inject domain-specific knowledge in parallel. Then we adopt a mixture-of-adapters gate to fuse the knowledge from different domain adapters dynamically. Our proposed Mixture-of-Domain-Adapters (MixDA) employs a two-stage adapter-tuning strategy that leverages both unlabeled data and labeled data to help the domain adaptation: i) domain-specific adapter on unlabeled data; followed by ii) the task-specific adapter on labeled data. MixDA can be seamlessly plugged into the pretraining-finetuning paradigm and our experiments demonstrate that MixDA achieves superior performance on in-domain tasks (GLUE), out-of-domain tasks (ChemProt, RCT, IMDB, Amazon), and knowledge-intensive tasks (KILT). Further analyses demonstrate the reliability, scalability, and efficiency of our method. The code is available at https://github.com/Amano-Aki/Mixture-of-Domain-Adapters.