Computation and Language (cs.CL)

Abstract: In most natural language inference problems, sentence representation is needed for semantic retrieval tasks. In recent years, pre-trained large language models have been quite effective for computing such representations. These models produce high-dimensional sentence embeddings. An evident performance gap between large and small models exists in practice. Hence, due to space and time hardware limitations, there is a need to attain comparable results when using the smaller model, which is usually a distilled version of the large language model. In this paper, we assess the model distillation of the sentence representation model Sentence-BERT by augmenting the pre-trained distilled model with a projection layer additionally learned on the Maximum Coding Rate Reduction (MCR2)objective, a novel approach developed for general-purpose manifold clustering. We demonstrate that the new language model with reduced complexity and sentence embedding size can achieve comparable results on semantic retrieval benchmarks.

Abstract: Language models have become nearly ubiquitous in natural language processing applications achieving state-of-the-art results in many tasks including prosody. As the model design does not define predetermined linguistic targets during training but rather aims at learning generalized representations of the language, analyzing and interpreting the representations that models implicitly capture is important in bridging the gap between interpretability and model performance. Several studies have explored the linguistic information that models capture providing some insights on their representational capacity. However, the current studies have not explored whether prosody is part of the structural information of the language that models learn. In this work, we perform a series of experiments on BERT probing the representations captured at different layers. Our results show that information about prosodic prominence spans across many layers but is mostly focused in middle layers suggesting that BERT relies mostly on syntactic and semantic information.

Abstract: Citations in scientific papers not only help us trace the intellectual lineage but also are a useful indicator of the scientific significance of the work. Citation intents prove beneficial as they specify the role of the citation in a given context. In this paper, we present CitePrompt, a framework which uses the hitherto unexplored approach of prompt-based learning for citation intent classification. We argue that with the proper choice of the pretrained language model, the prompt template, and the prompt verbalizer, we can not only get results that are better than or comparable to those obtained with the state-of-the-art methods but also do it with much less exterior information about the scientific document. We report state-of-the-art results on the ACL-ARC dataset, and also show significant improvement on the SciCite dataset over all baseline models except one. As suitably large labelled datasets for citation intent classification can be quite hard to find, in a first, we propose the conversion of this task to the few-shot and zero-shot settings. For the ACL-ARC dataset, we report a 53.86% F1 score for the zero-shot setting, which improves to 63.61% and 66.99% for the 5-shot and 10-shot settings, respectively.

Abstract: Currently, pre-trained language models (PLMs) do not cope well with the distribution shift problem, resulting in models trained on the training set failing in real test scenarios. To address this problem, the test-time adaptation (TTA) shows great potential, which updates model parameters to suit the test data at the testing time. Existing TTA methods rely on well-designed auxiliary tasks or self-training strategies based on pseudo-label. However, these methods do not achieve good trade-offs regarding performance gains and computational costs. To obtain some insights into such a dilemma, we take two representative TTA methods, i.e., Tent and OIL, for exploration and find that stable prediction is the key to achieving a good balance. Accordingly, in this paper, we propose perturbation consistency learning (PCL), a simple test-time adaptation method to promote the model to make stable predictions for samples with distribution shifts. Extensive experiments on adversarial robustness and cross-lingual transferring demonstrate that our method can achieve higher or comparable performance with less inference time over strong PLM backbones and previous state-of-the-art TTA methods.

Abstract: Structured State Spaces for Sequences (S4) is a recently proposed sequence model with successful applications in various tasks, e.g. vision, language modeling, and audio. Thanks to its mathematical formulation, it compresses its input to a single hidden state, and is able to capture long range dependencies while avoiding the need for an attention mechanism. In this work, we apply S4 to Machine Translation (MT), and evaluate several encoder-decoder variants on WMT'14 and WMT'16. In contrast with the success in language modeling, we find that S4 lags behind the Transformer by approximately 4 BLEU points, and that it counter-intuitively struggles with long sentences. Finally, we show that this gap is caused by S4's inability to summarize the full source sentence in a single hidden state, and show that we can close the gap by introducing an attention mechanism.

Abstract: Shannon Information theory has achieved great success in not only communication technology where it was originally developed for but also many other science and engineering fields such as machine learning and artificial intelligence. Inspired by the famous weighting scheme TF-IDF, we discovered that information entropy has a natural dual. We complement the classical Shannon information theory by proposing a novel quantity, namely troenpy. Troenpy measures the certainty, commonness and similarity of the underlying distribution. To demonstrate its usefulness, we propose a troenpy based weighting scheme for document with class labels, namely positive class frequency (PCF). On a collection of public datasets we show the PCF based weighting scheme outperforms the classical TF-IDF and a popular Optimal Transportation based word moving distance algorithm in a kNN setting. We further developed a new odds-ratio type feature, namely Expected Class Information Bias(ECIB), which can be regarded as the expected odds ratio of the information quantity entropy and troenpy. In the experiments we observe that including the new ECIB features and simple binary term features in a simple logistic regression model can further significantly improve the performance. The simple new weighting scheme and ECIB features are very effective and can be computed with linear order complexity.

Abstract: Many Natural Language Processing (NLP) systems use annotated corpora for training and evaluation. However, labeled data is often costly to obtain and scaling annotation projects is difficult, which is why annotation tasks are often outsourced to paid crowdworkers. Citizen Science is an alternative to crowdsourcing that is relatively unexplored in the context of NLP. To investigate whether and how well Citizen Science can be applied in this setting, we conduct an exploratory study into engaging different groups of volunteers in Citizen Science for NLP by re-annotating parts of a pre-existing crowdsourced dataset. Our results show that this can yield high-quality annotations and attract motivated volunteers, but also requires considering factors such as scalability, participation over time, and legal and ethical issues. We summarize lessons learned in the form of guidelines and provide our code and data to aid future work on Citizen Science.

Abstract: In this paper, we propose a methodology for task 10 of SemEval23, focusing on detecting and classifying online sexism in social media posts. The task is tackling a serious issue, as detecting harmful content on social media platforms is crucial for mitigating the harm of these posts on users. Our solution for this task is based on an ensemble of fine-tuned transformer-based models (BERTweet, RoBERTa, and DeBERTa). To alleviate problems related to class imbalance, and to improve the generalization capability of our model, we also experiment with data augmentation and semi-supervised learning. In particular, for data augmentation, we use back-translation, either on all classes, or on the underrepresented classes only. We analyze the impact of these strategies on the overall performance of the pipeline through extensive experiments. while for semi-supervised learning, we found that with a substantial amount of unlabelled, in-domain data available, semi-supervised learning can enhance the performance of certain models. Our proposed method (for which the source code is available on Github attains an F1-score of 0.8613 for sub-taskA, which ranked us 10th in the competition

Abstract: Evidence-aware fake news detection aims to conduct reasoning between news and evidence, which is retrieved based on news content, to find uniformity or inconsistency. However, we find evidence-aware detection models suffer from biases, i.e., spurious correlations between news/evidence contents and true/fake news labels, and are hard to be generalized to Out-Of-Distribution (OOD) situations. To deal with this, we propose a novel Dual Adversarial Learning (DAL) approach. We incorporate news-aspect and evidence-aspect debiasing discriminators, whose targets are both true/fake news labels, in DAL. Then, DAL reversely optimizes news-aspect and evidence-aspect debiasing discriminators to mitigate the impact of news and evidence content biases. At the same time, DAL also optimizes the main fake news predictor, so that the news-evidence interaction module can be learned. This process allows us to teach evidence-aware fake news detection models to better conduct news-evidence reasoning, and minimize the impact of content biases. To be noted, our proposed DAL approach is a plug-and-play module that works well with existing backbones. We conduct comprehensive experiments under two OOD settings, and plug DAL in four evidence-aware fake news detection backbones. Results demonstrate that, DAL significantly and stably outperforms the original backbones and some competitive debiasing methods.

Abstract: Human language is full of compositional syntactic structures, and although neural networks have contributed to groundbreaking improvements in computer systems that process language, widely-used neural network architectures still exhibit limitations in their ability to process syntax. To address this issue, prior work has proposed adding stack data structures to neural networks, drawing inspiration from theoretical connections between syntax and stacks. However, these methods employ deterministic stacks that are designed to track one parse at a time, whereas syntactic ambiguity, which requires a nondeterministic stack to parse, is extremely common in language. In this dissertation, we remedy this discrepancy by proposing a method of incorporating nondeterministic stacks into neural networks. We develop a differentiable data structure that efficiently simulates a nondeterministic pushdown automaton, representing an exponential number of computations with a dynamic programming algorithm. We incorporate this module into two predominant architectures: recurrent neural networks (RNNs) and transformers. We show that this raises their formal recognition power to arbitrary context-free languages, and also aids training, even on deterministic context-free languages. Empirically, neural networks with nondeterministic stacks learn context-free languages much more effectively than prior stack-augmented models, including a language with theoretically maximal parsing difficulty. We also show that an RNN augmented with a nondeterminsitic stack is capable of surprisingly powerful behavior, such as learning cross-serial dependencies, a well-known non-context-free pattern. We demonstrate improvements on natural language modeling and provide analysis on a syntactic generalization benchmark. This work represents an important step toward building systems that learn to use syntax in more human-like fashion.

Abstract: It is well-known that document context is vital for resolving a range of translation ambiguities, and in fact the document setting is the most natural setting for nearly all translation. It is therefore unfortunate that machine translation -- both research and production -- largely remains stuck in a decades-old sentence-level translation paradigm. It is also an increasingly glaring problem in light of competitive pressure from large language models, which are natively document-based. Much work in document-context machine translation exists, but for various reasons has been unable to catch hold. This paper suggests a path out of this rut by addressing three impediments at once: what architectures should we use? where do we get document-level information for training them? and how do we know whether they are any good? In contrast to work on specialized architectures, we show that the standard Transformer architecture is sufficient, provided it has enough capacity. Next, we address the training data issue by taking document samples from back-translated data only, where the data is not only more readily available, but is also of higher quality compared to parallel document data, which may contain machine translation output. Finally, we propose generative variants of existing contrastive metrics that are better able to discriminate among document systems. Results in four large-data language pairs (DE$\rightarrow$EN, EN$\rightarrow$DE, EN$\rightarrow$FR, and EN$\rightarrow$RU) establish the success of these three pieces together in improving document-level performance.

Abstract: This report describes GMU's sentiment analysis system for the SemEval-2023 shared task AfriSenti-SemEval. We participated in all three sub-tasks: Monolingual, Multilingual, and Zero-Shot. Our approach uses models initialized with AfroXLMR-large, a pre-trained multilingual language model trained on African languages and fine-tuned correspondingly. We also introduce augmented training data along with original training data. Alongside finetuning, we perform phylogeny-based adapter tuning to create several models and ensemble the best models for the final submission. Our system achieves the best F1-score on track 5: Amharic, with 6.2 points higher F1-score than the second-best performing system on this track. Overall, our system ranks 5th among the 10 systems participating in all 15 tracks.

Abstract: With increasing demand for and adoption of virtual assistants, recent work has investigated ways to accelerate bot schema design through the automatic induction of intents or the induction of slots and dialogue states. However, a lack of dedicated benchmarks and standardized evaluation has made progress difficult to track and comparisons between systems difficult to make. This challenge track, held as part of the Eleventh Dialog Systems Technology Challenge, introduces a benchmark that aims to evaluate methods for the automatic induction of customer intents in a realistic setting of customer service interactions between human agents and customers. We propose two subtasks for progressively tackling the automatic induction of intents and corresponding evaluation methodologies. We then present three datasets suitable for evaluating the tasks and propose simple baselines. Finally, we summarize the submissions and results of the challenge track, for which we received submissions from 34 teams.

Abstract: The development of large-scale Chinese language models is flourishing, yet there is a lack of corresponding capability assessments. Therefore, we propose a test to measure the multitask accuracy of large Chinese language models. This test encompasses four major domains, including medicine, law, psychology, and education, with 15 subtasks in medicine and 8 subtasks in education. We found that the best-performing models in the zero-shot setting outperformed the worst-performing models by nearly 22 percentage points on average. Across the four major domains, the average zero-shot accuracy of all models did not exceed 0.5. In the subdomains, only the GPT-3.5-turbo model achieved a zero-shot accuracy of 0.703 in clinical medicine, which was the highest accuracy among all models across all subtasks. All models performed poorly in the legal domain, with the highest zero-shot accuracy reaching only 0.259. By comprehensively evaluating the breadth and depth of knowledge across multiple disciplines, this test can more accurately identify the shortcomings of the models.

Abstract: Large language models (LLMs) have exhibited remarkable capabilities across a variety of domains and tasks, challenging our understanding of learning and cognition. Despite the recent success, current LLMs are not capable of processing complex audio information or conducting spoken conversations (like Siri or Alexa). In this work, we propose a multi-modal AI system named AudioGPT, which complements LLMs (i.e., ChatGPT) with 1) foundation models to process complex audio information and solve numerous understanding and generation tasks; and 2) the input/output interface (ASR, TTS) to support spoken dialogue. With an increasing demand to evaluate multi-modal LLMs of human intention understanding and cooperation with foundation models, we outline the principles and processes and test AudioGPT in terms of consistency, capability, and robustness. Experimental results demonstrate the capabilities of AudioGPT in solving AI tasks with speech, music, sound, and talking head understanding and generation in multi-round dialogues, which empower humans to create rich and diverse audio content with unprecedented ease. Our system is publicly available at \url{https://github.com/AIGC-Audio/AudioGPT}.

Abstract: Despite significant progress in Natural Language Generation for Indian languages (IndicNLP), there is a lack of datasets around complex structured tasks such as semantic parsing. One reason for this imminent gap is the complexity of the logical form, which makes English to multilingual translation difficult. The process involves alignment of logical forms, intents and slots with translated unstructured utterance. To address this, we propose an Inter-bilingual Seq2seq Semantic parsing dataset IE-SEMPARSE for 11 distinct Indian languages. We highlight the proposed task's practicality, and evaluate existing multilingual seq2seq models across several train-test strategies. Our experiment reveals a high correlation across performance of original multilingual semantic parsing datasets (such as mTOP, multilingual TOP and multiATIS++) and our proposed IE-SEMPARSE suite.

Abstract: Modern systems for multi-hop question answering (QA) typically break questions into a sequence of reasoning steps, termed chain-of-thought (CoT), before arriving at a final answer. Often, multiple chains are sampled and aggregated through a voting mechanism over the final answers, but the intermediate steps themselves are discarded. While such approaches improve performance, they do not consider the relations between intermediate steps across chains and do not provide a unified explanation for the predicted answer. We introduce Multi-Chain Reasoning (MCR), an approach which prompts large language models to meta-reason over multiple chains of thought, rather than aggregating their answers. MCR examines different reasoning chains, mixes information between them and selects the most relevant facts in generating an explanation and predicting the answer. MCR outperforms strong baselines on 7 multi-hop QA datasets. Moreover, our analysis reveals that MCR explanations exhibit high quality, enabling humans to verify its answers.