_id stringlengths 4 10 | text stringlengths 0 18.4k | title stringlengths 0 8.56k |
|---|---|---|
d226283502 | This paper seeks to uncover patterns of sound change across Indo-Aryan languages using an LSTM encoder-decoder architecture. We augment our models with embeddings representing language ID, part of speech, and other features such as word embeddings. We find that a highly augmented model shows highest accuracy in predicting held-out forms, and investigate other properties of interest learned by our models' representations. We outline extensions to this architecture that can better capture variation in Indo-Aryan sound change. | Disentangling dialects: a neural approach to Indo-Aryan historical phonology and subgrouping |
d9121717 | Commentary on Daelemans, Gillis, and Durieux | |
d2223811 | In modern biology, digitization of biosystematics publications is an important task. Extraction of taxonomic names from such documents is one of its major issues. This is because these names identify the various genera and species. This article reports on our experiences with learning techniques for this particular task. We say why established Named-Entity Recognition techniques are somewhat difficult to use in our context. One reason is that we have only very little training data available. Our experiments show that a combining approach that relies on regular expressions, heuristics, and word-level language recognition achieves very high precision and recall and allows to cope with those difficulties. | The Difficulties of Taxonomic Name Extraction and a Solution |
d51875175 | This paper reports the results of our transliteration experiments conducted on NEWS 2018 Shared Task dataset. We focus on creating the baseline systems trained using two open-source, statistical transliteration tools, namely Sequitur and Moses. We discuss the pre-processing steps performed on this dataset for both the systems. We also provide a re-ranking system which uses top hypotheses from Sequitur and Moses to create a consolidated list of transliterations. The results obtained from each of these models can be used to present a good starting point for the participating teams. | Statistical Machine Transliteration Baselines for NEWS 2018 |
d258841105 | Logical reasoning over incomplete knowledge graphs to answer complex logical queries is a challenging task. With the emergence of new entities and relations in constantly evolving KGs, inductive logical reasoning over KGs has become a crucial problem. However, previous PLMs-based methods struggle to model the logical structures of complex queries, which limits their ability to generalize within the same structure. In this paper, we propose a structuremodeled textual encoding framework for inductive logical reasoning over KGs. It encodes linearized query structures and entities using pre-trained language models to find answers. For structure modeling of complex queries, we design stepwise instructions that implicitly prompt PLMs on the execution order of geometric operations in each query. We further separately model different geometric operations (i.e., projection, intersection, and union) on the representation space using a pre-trained encoder with additional attention and maxout layers to enhance structured modeling. We conduct experiments on two inductive logical reasoning datasets and three transductive datasets. The results demonstrate the effectiveness of our method on logical reasoning over KGs in both inductive and transductive settings. | Query Structure Modeling for Inductive Logical Reasoning Over Knowledge Graphs |
d6746048 | Word-to-word dependency structures are useful for consistent representation and comparable evaluation of parsing results. However, most large-scale treebanks contain various variants of phrase structure trees, since automatic parsers usually produce constituent structures. We present a freely available extensible tool for converting phrase structure to dependencies automatically, and discuss its application to the NEGRA treebank of German. | Automatic transformation of phrase treebanks to dependency trees |
d44075965 | We take a multi-task learning approach to the shared Task 1 at SemEval-2018. The general idea concerning the model structure is to use as little external data as possible in order to preserve the task relatedness and reduce complexity. We employ multi-task learning with hard parameter sharing to exploit the relatedness between sub-tasks. As a base model, we use a standard recurrent neural network for both the classification and regression subtasks. Our system ranks 32nd out of 48 participants with a Pearson score of 0.557 in the first subtask, and 20th out of 35 in the fifth subtask with an accuracy score of 0.464. | KU-MTL at SemEval-2018 Task 1: Multi-task Identification of Affect in Tweets |
d220060833 | ||
d245853756 | We introduce CVSS, a massively multilingual-to-English speech-to-speech translation (S2ST) corpus, covering sentence-level parallel S2ST pairs from 21 languages into English. CVSS is derived from the Common Voice (Ardila et al., 2020) speech corpus and the CoVoST 2 (Wang et al., 2021b) speech-to-text translation (ST) corpus, by synthesizing the translation text from CoVoST 2 into speech using state-of-the-art TTS systems. Two versions of translation speech in English are provided: 1) CVSS-C: All the translation speech is in a single high-quality canonical voice; 2) CVSS-T: The translation speech is in voices transferred from the corresponding source speech. In addition, CVSS provides normalized translation text which matches the pronunciation in the translation speech. On each version of CVSS, we built baseline multilingual direct S2ST models and cascade S2ST models, verifying the effectiveness of the corpus. To build strong cascade S2ST baselines, we trained an ST model on CoVoST 2, which outperforms the previous state-of-the-art trained on the corpus without extra data by 5.8 BLEU. Nevertheless, the performance of the direct S2ST models approaches the strong cascade baselines when trained from scratch, and with only 0.1 or 0.7 BLEU difference on ASR transcribed translation when initialized from matching ST models. | CVSS Corpus and Massively Multilingual Speech-to-Speech Translation |
d6833074 | Recent research suggests that sentence structure can improve the accuracy of recognizing textual entailments and paraphrasing.Although background knowledge such as gazetteers, WordNet and custom built knowledge bases are also likely to improve performance, our goal in this paper is to characterize the syntactic features alone that aid in accurate entailment prediction. We describe candidate features, the role of machine learning, and two final decision rules. These rules resulted in an accuracy of 60.50 and 65.87% and average precision of 58.97 and 60.96% in RTE3 Test and suggest that sentence structure alone can improve entailment accuracy by 9.25 to 14.62% over the baseline majority class. | The Role of Sentence Structure in Recognizing Textual Entailment |
d227905489 | Traditional biomedical version of embeddings obtained from pre-trained language models have recently shown state-of-the-art results for relation extraction (RE) tasks in the medical domain.In this paper, we explore how to incorporate domain knowledge, available in the form of molecular structure of drugs, for predicting Drug-Drug Interaction from textual corpus. We propose a method, BERTChem-DDI, to efficiently combine drug embeddings obtained from the rich chemical structure of drugs (encoded in SMILES) along with off-the-shelf domain-specific BioBERT embedding-based RE architecture. Experiments conducted on the DDIExtraction 2013 corpus clearly indicate that this strategy improves other strong baselines architectures by 3.4% macro F1-score. | BERTChem-DDI : Improved Drug-Drug Interaction Prediction from text using Chemical Structure Information |
d253244423 | The research field of Legal Natural Language Processing (NLP) has been very active recently, with Legal Judgment Prediction (LJP) becoming one of the most extensively studied tasks. To date, most publicly released LJP datasets originate from countries with civil law. In this work, we release, for the first time, a challenging LJP dataset focused on class action cases in the US. It is the first dataset in the common law system that focuses on the harder and more realistic task involving the complaints as input instead of the often used facts summary written by the court. Additionally, we study the difficulty of the task by collecting expert human predictions, showing that even human experts can only reach 53% accuracy on this dataset. Our Longformer model clearly outperforms the human baseline (63%), despite only considering the first 2,048 tokens. Furthermore, we perform a detailed error analysis and find that the Longformer model is significantly better calibrated than the human experts. Finally, we publicly release the dataset and the code used for the experiments. | ClassActionPrediction: A Challenging Benchmark for Legal Judgment Prediction of Class Action Cases in the US |
d220045465 | In this paper, we present CorefQA, an accurate and extensible approach for the coreference resolution task. We formulate the problem as a span prediction task, like in question answering: A query is generated for each candidate mention using its surrounding context, and a span prediction module is employed to extract the text spans of the coreferences within the document using the generated query. This formulation comes with the following key advantages: (1) The span prediction strategy provides the flexibility of retrieving mentions left out at the mention proposal stage; (2) In the question answering framework, encoding the mention and its context explicitly in a query makes it possible to have a deep and thorough examination of cues embedded in the context of coreferent mentions; and (3) A plethora of existing question answering datasets can be used for data augmentation to improve the model's generalization capability. Experiments demonstrate significant performance boost over previous models, with 83.1 (+3.5) F1 score on the CoNLL-2012 benchmark and 87.5 (+2.5) F1 score on the GAP benchmark. 1 | CorefQA: Coreference Resolution as Query-based Span Prediction |
d237433537 | People convey their intention and attitude through linguistic styles of the text that they write. In this study, we investigate lexicon usages across styles throughout two lenses: human perception and machine word importance, since words differ in the strength of the stylistic cues that they provide. To collect labels of human perception, we curate a new dataset, HUMMINGBIRD, on top of benchmarking style datasets. We have crowd workers highlight the representative words in the text that makes them think the text has the following styles: politeness, sentiment, offensiveness, and five emotion types. We then compare these human word labels with word importance derived from a popular fine-tuned style classifier like BERT. Our results show that the BERT often finds content words not relevant to the target style as important words used in style prediction, but humans do not perceive the same way even though for some styles (e.g., positive sentiment and joy) humanand machine-identified words share significant overlap for some styles. | Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica |
d252780080 | Powerful generative models have led to recent progress in question generation (QG). However, it is difficult to measure advances in QG research since there are no standardized resources that allow a uniform comparison among approaches. In this paper, we introduce QG-Bench, a multilingual and multidomain benchmark for QG that unifies existing question answering datasets by converting them to a standard QG setting. It includes generalpurpose datasets such as SQuAD(Rajpurkar et al., 2016)for English, datasets from ten domains and two styles, as well as datasets in eight different languages. Using QG-Bench as a reference, we perform an extensive analysis of the capabilities of language models for the task. First, we propose robust QG baselines based on fine-tuning generative language models. Then, we complement automatic evaluation based on standard metrics with an extensive manual evaluation, which in turn sheds light on the difficulty of evaluating QG models. Finally, we analyse both the domain adaptability of these models as well as the effectiveness of multilingual models in languages other than English. QG-Bench is released along with the fine-tuned models presented in the paper, 1 which are also available as a demo. 2 | Generative Language Models for Paragraph-Level Question Generation |
d2685018 | This paper reports on the development of the prototype African Wordnet (AWN) which currently includes four languages. The resource has been developed by translating Common Base Concepts from English, and currently holds roughly 42 000 synsets. We describe here how some language specific and technical challenges have been overcome and discuss efforts to localise the content of the wordnet and quality assurance methods. A comparison of the number of synsets per language is given before concluding with plans to fast-track the development and for dissemination of the resource. | Taking stock of the African Wordnets project: 5 years of development |
d258378175 | Despite the success of Transformer models in vision and language tasks, they often learn knowledge from enormous data implicitly and cannot utilize structured input data directly. On the other hand, structured learning approaches such as graph neural networks (GNNs) that integrate prior information can barely compete with Transformer models. In this work, we aim to benefit from both worlds and propose a novel Multimodal Graph Transformer for question answering tasks that requires performing reasoning across multiple modalities. We introduce a graph-involved plug-and-play quasi-attention mechanism to incorporate multimodal graph information, acquired from text and visual data, to the vanilla self-attention as effective prior. In particular, we construct the text graph, dense region graph, and semantic graph to generate adjacency matrices, and then compose them with input vision and language features to perform downstream reasoning. Such a way of regularizing self-attention with graph information significantly improves the inferring ability and helps align features from different modalities. We validate the effectiveness of Multimodal Graph Transformer over its Transformer baselines on GQA, VQAv2, and MultiModalQA datasets. | Multimodal Graph Transformer for Multimodal Question Answering |
d248970443 | In the commercial aviation domain, there are a large number of documents, like accident reports of NTSB and ASRS, and regulatory directives ADs. There is a need for a system to efficiently access these diverse repositories to serve the demands of the aviation industry, such as maintenance, compliance, and safety. In this paper, we propose a Knowledge Graph (KG) guided Deep Learning (DL) based Question Answering (QA) system to cater to these requirements. We construct a KG from aircraft accident reports and contribute this resource to the community of researchers. The efficacy of this resource is tested and proved by the proposed QA system. Questions in Natural Language are converted into SPARQL (the interface language of the RDF graph database) queries and are answered from the KG. On the DL side, we examine two different QA models, BERT-QA and GPT3-QA, covering the two paradigms of answer formulation in QA. We evaluate our system on a set of handcrafted queries curated from the accident reports. Our hybrid KG + DL QA system, KGQA + BERT-QA, achieves 7% and 40.3% increase in accuracy over KGQA and BERT-QA systems respectively. Similarly, the other combined system, KGQA + GPT3-QA, achieves 29.3% and 9.3% increase in accuracy over KGQA and GPT3-QA systems respectively. Thus, we infer that the combination of KG and DL is better than either KG or DL individually for QA, at least in our chosen domain. | Knowledge Graph -Deep Learning: A Case Study in Question Answering in Aviation Safety Domain |
d3017040 | A web search with double checking model is proposed to explore the web as a live corpus. Five association measures including variants of Dice, Overlap Ratio, Jaccard, and Cosine, as well as Co-Occurrence Double Check (CODC), are presented. In the experiments on Rubenstein-Goodenough's benchmark data set, the CODC measure achieves correlation coefficient 0.8492, which competes with the performance (0.8914) of the model using WordNet. The experiments on link detection of named entities using the strategies of direct association, association matrix and scalar association matrix verify that the double-check frequencies are reliable. Further study on named entity clustering shows that the five measures are quite useful. In particular, CODC measure is very stable on wordword and name-name experiments. The application of CODC measure to expand community chains for personal name disambiguation achieves 9.65% and 14.22% increase compared to the system without community expansion. All the experiments illustrate that the novel model of web search with double checking is feasible for mining associations from the web. | Novel Association Measures Using Web Search with Double Checking |
d254125751 | This work introduces a new multi-task, parameter-efficient language model (LM) tuning method that learns to transfer knowledge across different tasks via a mixture of soft prompts-small prefix embedding vectors pretrained for different tasks. Our method, called ATTEMPT (ATTEntional Mixtures of Prompt Tuning), obtains source prompts as encodings of large-scale source tasks into a small number of parameters and trains an attention module to interpolate the source prompts and a newly initialized target prompt for every instance in the target task. During training, only the target task prompt and the attention weights, which are shared between tasks in multi-task training, are updated, while the original LM and source prompts are intact. ATTEMPT is highly parameter-efficient (e.g., updates 2,300 times fewer parameters than full fine-tuning), while achieving high task performance using knowledge from high-resource tasks. Moreover, it is modular using pre-trained soft prompts and can flexibly add or remove source prompts for effective knowledge transfer. Our experimental results across 21 diverse NLP datasets show that ATTEMPT significantly outperforms prompt tuning and outperforms or matches fully finetuned or other parameter-efficient tuning approaches that use over ten times more parameters. Finally, ATTEMPT outperforms previous work in few-shot learning settings. 1 | ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Prompts |
d219305928 | ||
d251718718 | Knowledge graph (KG) inference aims to address the natural incompleteness of KGs, including rule learning-based and KG embedding (KGE) models. However, the rule learningbased models suffer from low efficiency and generalization while KGE models lack interpretability. To address these challenges, we propose a novel and effective closed-loop neuralsymbolic learning framework EngineKG via incorporating our developed KGE and rule learning modules. KGE module exploits symbolic rules and paths to enhance the semantic association between entities and relations for improving KG embeddings and interpretability. A novel rule pruning mechanism is proposed in the rule learning module by leveraging paths as initial candidate rules and employing KG embeddings together with concepts for extracting more high-quality rules. Experimental results on four real-world datasets show that our model outperforms the relevant baselines on link prediction tasks, demonstrating the superiority of our KG inference model in a neural-symbolic learning fashion. The source code and datasets of this paper are available at https://github.com/ngl567/EngineKG. | Perform Like an Engine: A Closed-Loop Neural-Symbolic Learning Framework for Knowledge Graph Inference |
d236486165 | ||
d256662602 | Recent transformer language models achieve outstanding results in many natural language processing (NLP) tasks. However, their enormous size often makes them impractical on memory-constrained devices, requiring practitioners to compress them to smaller networks. In this paper, we explore offline compression methods, meaning computationally-cheap approaches that do not require further finetuning of the compressed model. We challenge the classical matrix factorization methods by proposing a novel, better-performing autoencoder-based framework. We perform a comprehensive ablation study of our approach, examining its different aspects over a diverse set of evaluation settings. Moreover, we show that enabling collaboration between modules across layers by compressing certain modules together positively impacts the final model performance. Experiments on various NLP tasks demonstrate that our approach significantly outperforms commonly used factorization-based offline compression methods. 1 . Bourdev. 2014. Compressing deep convolutional networks using vector quantization. arXiv preprint arXiv:1412.6115. Nathan Halko, Per-Gunnar Martinsson, and Joel A Tropp. 2011. Finding structure with randomness: Probabilistic algorithms for constructing approximate matrix decompositions. SIAM review, 53(2):217-288. | Revisiting Offline Compression: Going Beyond Factorization-based Methods for Transformer Language Models |
d254274924 | Knowledge about outcomes is critical for complex event understanding but is hard to acquire. We show that by pre-identifying a participant in a complex event, crowdworkers are able to (1) infer the collective impact of salient events that make up the situation, (2) annotate the volitional engagement of participants in causing the situation, and (3) ground the outcome of the situation in state changes of the participants. By creating a multi-step interface and a careful quality control strategy, we collect a high quality annotated dataset of 8K short newswire narratives and ROCStories with high inter-annotator agreement (0.74-0.96 weighted Fleiss Kappa). Our dataset, POQue (Participant Outcome Questions), enables the exploration and development of models that address multiple aspects of semantic understanding. Experimentally, we show that current language models lag behind human performance in subtle ways through our task formulations that target abstract and specific comprehension of a complex event, its outcome, and a participant's influence over the event culmination. | POQue: Asking Participant-specific Outcome Questions for a Deeper Understanding of Complex Events |
d9737131 | In a sequence of papers, Moschovakis developed a class of languages of recursion as a new approach to the mathematical notion of algorithm and development of computational semantics, e.g., see Moschovakis [7], for FLR, and Moschovakis [8], for L λ ar . In particular, the language and theory of acyclic recursion L λ ar is intended for modeling the logical concepts of meaning and synonymy, from the perspective of the theory of computability, by targeting adequateness of computational semantics of NL. L λ ar is a higher order type theory, which is a proper extension of Gallin's TY 2 , Gallin [3], and, thus, of Montague's Intensional Logic (IL). L λ ar has a highly expressive language, an effective reduction calculus and strong mathematical properties. It models the notion of algorithm by abstract mathematical objects, which are tuple of functions defined by mutual recursion, called acyclic recursors. The referential intensions of the meaningful L λ ar terms are acyclic recursors defined by their canonical forms, which are recursion terms. For the construction of recursion terms (where-terms), the language L λ ar uses a recursion operator, denoted by the constant where that applies over a head term A 0 and a set of assignments, called body, {p 1 := A 1 , . . . , p n := A n }, where each A i is a term of the same type as the recursion variable p i (1 ≤ i ≤ n): A 0 where{p 1 := A 1 , . . . , p n := A n }. The where-terms represent recursive computations by designating functional recursors: intuitively, the denotation of the term A 0 depends on the functions denoted by p 1 , . . . , p n that are computed recursively by the system of assignments {p 1 := A 1 , . . . , p n := A n }. In an acyclic system of assignments, the computations close-off. The formal syntax of L λ ar allows only recursion terms with acyclic systems of assignments, while the FLR allows cyclicity, but is limited with respect to its type system. The languages of recursion (e.g., FLR and L λ ar ) have two semantic layers: denotational semantics and ref-295 | Formalisation of Intensionality as Algorithms 1 Background and Recent Development |
d259370565 | Despite the extensive applications of relation extraction (RE) tasks in various domains, little has been explored in the historical context, which contains promising data across hundreds and thousands of years. To promote the historical RE research, we present HistRED constructed from Yeonhaengnok. Yeonhaengnok is a collection of records originally written in Hanja, the classical Chinese writing, which has later been translated into Korean. HistRED provides bilingual annotations such that RE can be performed on Korean and Hanja texts. In addition, HistRED supports various self-contained subtexts with different lengths, from a sentence level to a document level, supporting diverse context settings for researchers to evaluate the robustness of their RE models. To demonstrate the usefulness of our dataset, we propose a bilingual RE model that leverages both Korean and Hanja contexts to predict relations between entities. Our model outperforms monolingual baselines on HistRED, showing that employing multiple language contexts supplements the RE predictions. The dataset is publicly available at: https://huggingface.co/ datasets/Soyoung/HistRED under CC BY-NC-ND 4.0 license.English* | HistRED: A Historical Document-Level Relation Extraction Dataset |
d259370722 | Automatic melody-to-lyric generation is a task in which song lyrics are generated to go with a given melody. It is of significant practical interest and more challenging than unconstrained lyric generation as the music imposes additional constraints onto the lyrics. The training data is limited as most songs are copyrighted, resulting in models that underfit the complicated cross-modal relationship between melody and lyrics. In this work, we propose a method for generating high-quality lyrics without training on any aligned melody-lyric data. Specifically, we design a hierarchical lyric generation framework that first generates a song outline and second the complete lyrics. The framework enables disentanglement of training (based purely on text) from inference (melodyguided text generation) to circumvent the shortage of parallel data.We leverage the segmentation and rhythm alignment between melody and lyrics to compile the given melody into decoding constraints as guidance during inference. The two-step hierarchical design also enables content control via the lyric outline, a much-desired feature for democratizing collaborative song creation. Experimental results show that our model can generate high-quality lyrics that are more on-topic, singable, intelligible, and coherent than strong baselines, for example SongMASS (Sheng et al., 2021), a SOTA model trained on a parallel dataset, with a 24% relative overall quality improvement based on human ratings. 1 | Unsupervised Melody-to-Lyric Generation |
d259370875 | Ambiguity is a major obstacle to providing services based on sentence classification. However, because of the structural limitations of the service, there may not be sufficient contextual information to resolve the ambiguity. In this situation, we focus on ambiguity detection so that service design considering ambiguity is possible. We utilize similarity in a semantic space to detect ambiguity in service scenarios 1 and training data. In addition, we apply taskspecific embedding to improve performance. Our results demonstrate that ambiguities and resulting labeling errors in training data or scenarios can be detected. Additionally, we confirm that it can be used to debug services. | Semantic Ambiguity Detection in Sentence Classification using Task-Specific Embeddings |
d225068144 | This paper describes the submission of LMU Munich to the WMT 2020 unsupervised shared task, in two language directions, German↔Upper Sorbian. Our core unsupervised neural machine translation (UNMT) system follows the strategy of Chronopoulou et al. (2020), using a monolingual pretrained language generation model (on German) and finetuning it on both German and Upper Sorbian, before initializing a UNMT model, which is trained with online backtranslation. Pseudoparallel data obtained from an unsupervised statistical machine translation (USMT) system is used to fine-tune the UNMT model. We also apply BPE-Dropout to the low-resource (Upper Sorbian) data to obtain a more robust system. We additionally experiment with residual adapters and find them useful in the Upper Sorbian→German direction. We explore sampling during backtranslation and curriculum learning to use SMT translations in a more principled way. Finally, we ensemble our bestperforming systems and reach a BLEU score of 32.4 on German→Upper Sorbian and 35.2 on Upper Sorbian→German. | The LMU Munich System for the WMT 2020 Unsupervised Machine Translation Shared Task |
d13424418 | We describe a general framework for encoding rich domain models and sophisticated plan reasoning capabilities. The approach uses graph-basedreasoning to address a wide range of tasks that typically arise in dialogue systems. The graphical plan representation is independent of but connected to the underlying representation of action and time. We describe types of plan recognition that are needed, illustrating these with examples from dialogues collected as part of the TRAINS project. The algorithms for the tasks are presented, and issues in the formalization of the reasoning processes are discussed. | Generic Plan Recognition for Dialogue Systems |
d257921487 | The Perso-Arabic scripts are a family of scripts that are widely adopted and used by various linguistic communities around the globe. Identifying various languages using such scripts is crucial to language technologies and challenging in low-resource setups. As such, this paper sheds light on the challenges of detecting languages using Perso-Arabic scripts, especially in bilingual communities where "unconventional" writing is practiced. To address this, we use a set of supervised techniques to classify sentences into their languages. Building on these, we also propose a hierarchical model that targets clusters of languages that are more often confused by the classifiers. Our experiment results indicate the effectiveness of our solutions. 1 | PALI: A Language Identification Benchmark for Perso-Arabic Scripts |
d248006116 | Topic models are some of the most popular ways to represent textual data in an interpretable manner. Recently, advances in deep generative models, specifically auto-encoding variational Bayes (AEVB), have led to the introduction of unsupervised neural topic models, which leverage deep generative models as opposed to traditional statistics-based topic models. We extend upon these neural topic models by introducing the Label-Indexed Neural Topic Model (LI-NTM), which is, to the extent of our knowledge, the first effective upstream semisupervised neural topic model. We find that LI-NTM outperforms existing neural topic models in document reconstruction benchmarks, with the most notable results in low labeled data regimes and for data-sets with informative labels; furthermore, our jointly learned classifier outperforms baseline classifiers in ablation studies. | A Joint Learning Approach for Semi-supervised Neural Topic Modeling |
d584190 | Over the last few decades, significant strides have been made in handwriting recognition (HR), which is the automatic transcription of handwritten documents. HR often focuses on modern handwritten material, but in the electronic age, the volume of handwritten material is rapidly declining. However, we believe HR is on the verge of having major application to historical record collections. In recent years, archives and genealogical organizations have conducted huge campaigns to transcribe valuable historical record content with such transcription being largely done through human-intensive labor. HR has the potential of revolutionizing these transcription endeavors. To test the hypothesis that this technology is close to applicability, and to provide a testbed for reducing any accuracy gaps, we have developed an evaluation paradigm for historical record handwriting recognition. We created a huge test corpus consisting of four historical data collections of four differing genres and three languages. In this paper, we provide the details of these extensive resources which we intend to release to the research community for further study. Since several research organizations have already participated in this evaluation, we also show initial results and comparisons to human levels of performance. | Corpus and Evaluation of Handwriting Recognition of Historical Genealogical Records |
d259138703 | Endowing chatbots with a consistent persona is essential to an engaging conversation, yet it remains an unresolved challenge. In this work, we propose a new retrieval-enhanced approach for personalized response generation. Specifically, we design a hierarchical transformer retriever trained on dialogue domain data to perform personalized retrieval and a context-aware prefix encoder that fuses the retrieved information to the decoder more effectively. Extensive experiments on a real-world dataset demonstrate the effectiveness of our model at generating more fluent and personalized responses. We quantitatively evaluate our model's performance under a suite of human and automatic metrics and find it to be superior compared to state-of-the-art baselines on English Reddit conversations. 1 | RECAP: Retrieval-Enhanced Context-Aware Prefix Encoder for Personalized Dialogue Response Generation |
d16500537 | This work describes how derivation tree fragments based on a variant of Tree Adjoining Grammar (TAG) can be used to check treebank consistency. Annotation of word sequences are compared both for their internal structural consistency, and their external relation to the rest of the tree. We expand on earlier work in this area in three ways. First, we provide a more complete description of the system, showing how a naive use of TAG structures will not work, leading to a necessary refinement. We also provide a more complete account of the processing pipeline, including the grouping together of structurally similar errors and their elimination of duplicates. Second, we include the new experimental external relation check to find an additional class of errors. Third, we broaden the evaluation to include both the internal and external relation checks, and evaluate the system on both an Arabic and English treebank. The evaluation has been successful enough that the internal check has been integrated into the standard pipeline for current English treebank construction at the Linguistic Data Consortium | Further Developments in Treebank Error Detection Using Derivation Trees |
d252763328 | Named entity recognition and relation extraction are core sub-tasks of relational triple extraction. Recent studies have used parameter sharing or joint decoding to create interaction between these two tasks. However, ensuring the specificity of task-specific traits while the two tasks interact properly is a huge difficulty. We propose a multi-gate encoder that models bidirectional task interaction while keeping sufficient feature specificity based on gating mechanism in this paper. Precisely, we design two types of independent gates: task gates to generate task-specific features and interaction gates to generate instructive features to guide the opposite task. Our experiments show that our method increases the state-of-the-art (SOTA) relation F1 scores on ACE04, ACE05 and SciERC datasets to 63. | A Multi-Gate Encoder for Joint Entity and Relation Extraction respectively, with higher inference speed over previous SOTA model |
d6568223 | In this paper we report on several issues arising out of a first attempt to annotate task-oriented spoken dialog for rhetorical structure using Rhetorical Structure Theory. We discuss an annotation scheme we are developing to resolve the difficulties we have encountered. | Rhetorical structure in dialog* |
d11335565 | Chinese word segmentation (CWS) is the fundamental technology for many NLPrelated applications. It is reported that more than 60% of segmentation errors is caused by the out-of-vocabulary (OOV) words. Recent studies in CWS show that, statistical machine learning method is, to some extent, effective on solving OOV words. But labeled data is limited in size and unbalanced in content which makes it impossible to obtain all the required knowledge to recognize OOV words. In this paper, large scaled web data is incorporated as knowledge supplement. A framework which combines using web search technology and machine learning method is proposed. For each sentence, basic segmentation is performed using linear-chain Conditional Random Fields (CRF) model. Substrings which CRF model gives low confidence decisions are extracted and sent to search engine to perform web search based word segmentation. Final decision is made by considering both CRF model based segmentation result and that of web search based result. Evaluations are conducted on SIGHAN Bakeoff 2005 and 2006 datasets, showing the effectiveness of the proposed framework on dealing with OOV words. | Incorporate Web Search Technology to Solve Out-of-Vocabulary Words in Chinese Word Segmentation |
d14908610 | This paper presents a novel semisupervised learning algorithm called Active Deep Networks (ADN), to address the semi-supervised sentiment classification problem with active learning. First, we propose the semi-supervised learning method of ADN. ADN is constructed by Restricted Boltzmann Machines (RBM) with unsupervised learning using labeled data and abundant of unlabeled data. Then the constructed structure is finetuned by gradient-descent based supervised learning with an exponential loss function. Second, we apply active learning in the semi-supervised learning framework to identify reviews that should be labeled as training data. Then ADN architecture is trained by the selected labeled data and all unlabeled data. Experiments on five sentiment classification datasets show that ADN outperforms the semi-supervised learning algorithm and deep learning techniques applied for sentiment classification. | Active Deep Networks for Semi-Supervised Sentiment Classification |
d16344167 | In this paper, we present the methodological principles and the implementation framework of text annotation process in an Information Extraction setting. Due to the recent prevalence of XML as a means for describing structured documents in a reusable format, our team has switched to an XML based annotation schema. In that framework, an XML annotation platform has been built, while processing tools, lexical resources and textual data communicate with each other via this platform. Editing/viewing tools have been implemented, endowed with functionalities that allow annotators to gain access to previous annotation levels as well as necessary lexical resources. | Multi-level XML-based Corpus Annotation |
d40013876 | Dans cet article, nous proposons une approche de conversion graphème-phonème pour les noms propres. L'approche repose sur une méthode probabiliste : les Champs Aléatoires Conditionnels (Conditional Random Fields, CRF). Les CRFs donnent une prévision à long terme, n'exigent pas l'indépendance des observations et permettent l'intégration de tags. Dans nos travaux antérieurs, l'approche de conversion graphème-phonème utilisant les CRFs a été proposée pour les mots communs et différents paramétrages des CRFs ont été étudiés. Dans cet article, nous étendons ce travail aux noms propres. Par ailleurs, nous proposons un algorithme pour la détection de l'origine des noms propres. Le système proposé est validé sur deux dictionnaires de prononciations. Notre approche se compare favorablement aux JMM (Joint-Multigram Model, système de l'état de l'art), et tire profit de la connaissance de la langue d'origine du nom propre.AbstractPronunciation generation for proper names using Conditional Random FieldsWe propose an approach to grapheme-to-phoneme conversion for proper names based on a probabilistic method: Conditional Random Fields (CRFs). CRFs give a long term prediction, assume a relafex state independence condition and allow a tag integration. In previous work, grapheme-to-phoneme conversion using CRF has been proposed for non proper names and different CRF features are studied. In this paper, we extend this work to proper names. Moreover, we propose an algorithm for origine detection of proper names of foreign origins. The proposed system is validated on two pronunciation dictionaries. Our approach compares favorably with the performance of the state-of-the-art Joint-Multigram Models and takes advantage of the knowledge of the origin of the proper name. | Génération des prononciations de noms propres à l'aide des Champs Aléatoires Conditionnels |
d7450499 | This paper describes our participation in BUCC 2017 shared task: identifying parallel sentences in comparable corpora. Our goal is to leverage continuous vector representations and distributional semantics with a minimal use of external preprocessing and postprocessing tools. We report experiments that were conducted after transmitting our results. | Framework for Identifying Parallel Sentences in Comparable Corpora |
d2868925 | The paper discusses a recent extension of the linguistic framework of the Rosetta system. The original framework is elegant and has proved its value in practice, but it also has a number of deficiencies, of which the most salient is the impossibility to assign an explicit structure to the grammars. This may cause problems, especially in a situation where large grammars have to be written by a group of people. The newly developed framework enables us to divide a grammar into subgrammars in a linguistically motivated way and to control explicitly the application of rules in a subgrammar. On the other hand it enables us to divide the set of grammar rules into rule classes in such a way that we get hold of the more difficult translation relations. The use of both these divisions naturally leads to a highly modular structure of the system, which helps in controlling its complexity. We will show that these divisions also give insight into a class of difficult translation problems in which there is a mismatch of categories. | Subgrammars, Rule Classes and Control in the Rosetta Translation System * |
d7502112 | This paper describes automatic techniques for mapping 9611 entries in a database of English verbs to Word-Net senses. The verbs were initially grouped into 491 classes based on syntactic features.Mapping these verbs into WordNet senses provides a resource that supports disambiguation in multilingual applications such as machine translation and cross-language information retrieval. Our techniques make use of (1) a training set of 1791 disambiguated entries, representing 1442 verb entries from 167 classes;(2) word sense probabilities, from frequency counts in a tagged corpus;(3) semantic similarity of WordNet senses for verbs within the same class; (4) probabilistic correlations between WordNet data and attributes of the verb classes. The best results achieved 72% precision and 58% recall, versus a lower bound of 62% precision and 38% recall for assigning the most frequently occurring WordNet sense, and an upper bound of 87% precision and 75% recall for human judgment. ¨ S impleProd: Product of all simple measures SimpleWtdSum: Weighted sum of all simple measures MajSimpleSgl: Majority vote of all (7) simple voters MajSimplePair: Majority vote of all (21) pairs of simple voters 8 | Mapping Lexical Entries in a Verbs Database to WordNet Senses |
d245334944 | Developing Natural Language Processing re sources for a low resource language is a chal lenging but essential task. In this paper, we present a Morphological Analyzer for Gujarati. We have used a BiDirectional LSTM based approach to perform morpheme boundary de tection and grammatical feature tagging. We have created a data set of Gujarati words with lemma and grammatical features. The Bi LSTM based model of Morph Analyzer dis cussed in the paper handles the language mor phology effectively without the knowledge of any handcrafted suffix rules. To the best of our knowledge, this is the first dataset and morph analyzer model for the Gujarati lan guage which performs both grammatical fea ture tagging and morpheme boundary detec tion tasks. | Morpheme Boundary Detection & Grammatical Feature Prediction for Gujarati : Dataset & Model |
d239016866 | Responsing with image has been recognized as an important capability for an intelligent conversational agent. Yet existing works only focus on exploring the multimodal dialogue models which depend on retrieval-based methods, but neglecting generation methods. To fill in the gaps, we first present a new task: multimodal dialogue response generation (MDRG)given the dialogue context, one model needs to generate a text or an image as response. Learning such a MDRG model often requires multimodal dialogues containing both texts and images which are difficult to obtain. Motivated by the challenge in practice, we consider MDRG under a natural assumption that only limited training examples are available. Under such a low-resource setting, we devise a novel conversational agent, Divter, in order to isolate parameters that depend on multimodal dialogues from the entire generation model. By this means, the major part of the model can be learned from a large number of text-only dialogues and textimage pairs respectively, then the whole parameters can be well fitted using just a few training examples. Extensive experiments demonstrate our method achieves state-of-the-art results in both automatic and human evaluation, and can generate informative text and high-resolution image responses. | Multimodal Dialogue Response Generation |
d247315257 | The task of joint dialog sentiment classification (DSC) and act recognition (DAR) aims to simultaneously predict the sentiment label and act label for each utterance in a dialog. In this paper, we put forward a new framework which models the explicit dependencies via integrating prediction-level interactions other than semantics-level interactions, more consistent with human intuition. Besides, we propose a speaker-aware temporal graph (SATG) and a dual-task relational temporal graph (DRTG) to introduce temporal relations into dialog understanding and dual-task reasoning. To implement our framework, we propose a novel model dubbed DARER, which first generates the context-, speaker-and temporal-sensitive utterance representations via modeling SATG, then conducts recurrent dual-task relational reasoning on DRTG, in which process the estimated label distributions act as key clues in prediction-level interactions. Experiment results show that DARER outperforms existing models by large margins while requiring much less computation resource and costing less training time. Remarkably, on DSC task in Mastodon, DARER gains a relative improvement of about 25% over previous best model in terms of F1, with less than 50% parameters and about only 60% required GPU memory. | DARER: Dual-task Temporal Relational Recurrent Reasoning Network for Joint Dialog Sentiment Classification and Act Recognition |
d259095915 | Humor is a central aspect of human communication that has not been solved for artificial agents so far. Large language models (LLMs) are increasingly able to capture implicit and contextual information. Especially, OpenAI's ChatGPT recently gained immense public attention. The GPT3-based model almost seems to communicate on a human level and can even tell jokes. Humor is an essential component of human communication. But is ChatGPT really funny?We put ChatGPT's sense of humor to the test. In a series of exploratory experiments around jokes, i.e., generation, explanation, and detection, we seek to understand ChatGPT's capability to grasp and reproduce human humor. Since the model itself is not accessible, we applied prompt-based experiments.Our empirical evidence indicates that jokes are not hard-coded but mostly also not newly generated by the model. Over 90% of 1008 generated jokes were the same 25 Jokes. The system accurately explains valid jokes but also comes up with fictional explanations for invalid jokes. Joke-typical characteristics can mislead Chat-GPT in the classification of jokes. ChatGPT has not solved computational humor yet but it can be a big leap toward "funny" machines. | ChatGPT is fun, but it is not funny! Humor is still challenging Large Language Models |
d16606493 | This paper describes the characteristics and structure of a Basque singing voice database of bertsolaritza. Bertsolaritza is a popular singing style from Basque Country sung exclusively in Basque that is improvised and a capella. The database is designed to be used in statistical singing voice synthesis for bertsolaritza style. Starting from the recordings and transcriptions of numerous singers, diarization and phoneme alignment experiments have been made to extract the singing voice from the recordings and create phoneme alignments. This labelling processes have been performed applying standard speech processing techniques and the results prove that these techniques can be used in this specific singing style. | A singing voice database in Basque for statistical singing synthesis of bertsolaritza |
d207917676 | The field of question answering (QA) has seen rapid growth in new tasks and modeling approaches in recent years. Large scale datasets and focus on challenging linguistic phenomena have driven development in neural models, some of which have achieved parity with human performance in limited cases. However, an examination of state-of-the-art model output reveals that a gap remains in reasoning ability compared to a human, and performance tends to degrade when models are exposed to less-constrained tasks. We are interested in more clearly defining the strengths and limitations of leading models across diverse QA challenges, intending to help future researchers with identifying pathways to generalizable performance. We conduct extensive qualitative and quantitative analyses on the results of four models across four datasets and relate common errors to model capabilities. We also illustrate limitations in the datasets we examine and discuss a way forward for achieving generalizable models and datasets that broadly test QA capabilities. | Bend but Don't Break? Multi-Challenge Stress Test for QA Models |
d237364609 | Pretrained transformer-based models such as BERT have demonstrated state-of-the-art predictive performance when adapted into a range of natural language processing tasks. An open problem is how to improve the faithfulness of explanations (rationales) for the predictions of these models. In this paper, we hypothesize that salient information extracted a priori from the training data can complement the task-specific information learned by the model during fine-tuning on a downstream task. In this way, we aim to help BERT not to forget assigning importance to informative input tokens when making predictions by proposing SALOSS; an auxiliary loss function for guiding the multi-head attention mechanism during training to be close to salient information extracted a priori using TextRank. Experiments for explanation faithfulness across five datasets, show that models trained with SA-LOSS consistently provide more faithful explanations across four different feature attribution methods compared to vanilla BERT. Using the rationales extracted from vanilla BERT and SALOSS models to train inherently faithful classifiers, we further show that the latter result in higher predictive performance in downstream tasks. 1 | Enjoy the Salience: Towards Better Transformer-based Faithful Explanations with Word Salience |
d64403004 | ||
d258212810 | Despite much progress in recent years, the vast majority of work in natural language processing (NLP) is on standard languages with many speakers. In this work, we instead focus on low-resource languages and in particular non-standardized lowresource languages. Even within branches of major language families, often considered well-researched, little is known about the extent and type of available resources and what the major NLP challenges are for these language varieties. The first step to address this situation is a systematic survey of available corpora (most importantly, annotated corpora, which are particularly valuable for NLP research). Focusing on Germanic low-resource language varieties, we provide such a survey in this paper. Except for geolocation (origin of speaker or document), we find that manually annotated linguistic resources are sparse and, if they exist, mostly cover morphosyntax. Despite this lack of resources, we observe that interest in this area is increasing: there is active development and a growing research community. To facilitate research, we make our overview of over 80 corpora publicly available. 1 | A Survey of Corpora for Germanic Low-Resource Languages and Dialects |
d223953716 | The majority of work in targeted sentiment analysis has concentrated on finding better methods to improve the overall results. Within this paper we show that these models are not robust to linguistic phenomena, specifically negation and speculation. In this paper, we propose a multi-task learning method to incorporate information from syntactic and semantic auxiliary tasks, including negation and speculation scope detection, to create English-language models that are more robust to these phenomena.Further we create two challenge datasets to evaluate model performance on negated and speculative samples. We find that multi-task models and transfer learning via language modelling can improve performance on these challenge datasets, but the overall performances indicate that there is still much room for improvement. We release both the datasets and the source code at https://github.com/ jerbarnes/multitask_negation_ for_targeted_sentiment. | Multi-task Learning of Negation and Speculation for Targeted Sentiment Classification |
d29272352 | The relevance of syntactic dependency annotated corpora is nowadays unquestioned. However, a broad debate on the optimal set of dependency relation tags did not take place yet. As a result, largely varying tag sets of a largely varying size are used in different annotation initiatives. We propose a hierarchical dependency structure annotation schema that is more detailed and more flexible than the known annotation schemata. The schema allows us to choose the level of the desired detail of annotation, which facilitates the use of the schema for corpus annotation for different languages and for different NLP applications. Thanks to the inclusion of semanticosyntactic tags into the schema, we can annotate a corpus not only with syntactic dependency structures, but also with valency patterns as they are usually found in separate treebanks such as PropBank and NomBank. Semantico-syntactic tags and the level of detail of the schema furthermore facilitate the derivation of deep-syntactic and semantic annotations, leading to truly multilevel annotated dependency corpora. Such multilevel annotations can be readily used for the task of ML-based acquisition of grammar resources that map between the different levels of linguistic representation -something which forms part of, for instance, any natural language text generator. | Syntactic Dependencies for Multilingual and Multilevel Corpus Annotation |
d17445714 | We present a method for predicting machine translation output quality geared to the needs of computer-assisted translation. These include the capability to: i) continuously learn and self-adapt to a stream of data coming from multiple translation jobs, ii) react to data diversity by exploiting human feedback, and iii) leverage data similarity by learning and transferring knowledge across domains. To achieve these goals, we combine two supervised machine learning paradigms, online and multitask learning, adapting and unifying them in a single framework. We show the effectiveness of our approach in a regression task (HTER prediction), in which online multitask learning outperforms the competitive online single-task and pooling methods used for comparison. This indicates the feasibility of integrating in a CAT tool a single QE component capable to simultaneously serve (and continuously learn from) multiple translation jobs involving different domains and users. | Online Multitask Learning for Machine Translation Quality Estimation |
d247315450 | Neural networks tend to gradually forget the previously learned knowledge when learning multiple tasks sequentially from dynamic data distributions. This problem is called catastrophic forgetting, which is a fundamental challenge in the continual learning of neural networks. In this work, we observe that catastrophic forgetting not only occurs in continual learning but also affects the traditional static training. Neural networks, especially neural machine translation models, suffer from catastrophic forgetting even if they learn from a static training set. To be specific, the final model pays imbalanced attention to training samples, where recently exposed samples attract more attention than earlier samples. The underlying cause is that training samples do not get balanced training in each model update, so we name this problem imbalanced training. To alleviate this problem, we propose Complementary Online Knowledge Distillation (COKD), which uses dynamically updated teacher models trained on specific data orders to iteratively provide complementary knowledge to the student model. Experimental results on multiple machine translation tasks show that our method successfully alleviates the problem of imbalanced training and achieves substantial improvements over strong baseline systems. 1 | Overcoming Catastrophic Forgetting beyond Continual Learning: Balanced Training for Neural Machine Translation |
d11497074 | Distant supervision (DS) is an appealing learning method which learns from existing relational facts to extract more from a text corpus. However, the accuracy is still not satisfying. In this paper, we point out and analyze some critical factors in DS which have great impact on accuracy, including valid entity type detection, negative training examples construction and ensembles. We propose an approach to handle these factors. By experimenting on Wikipedia articles to extract the facts in Freebase (the top 92 relations), we show the impact of these three factors on the accuracy of DS and the remarkable improvement led by the proposed approach. | Towards Accurate Distant Supervision for Relational Facts Extraction |
d32575724 | Many unsupervised learning techniques have been proposed to obtain meaningful representations of words from text. In this study, we evaluate these various techniques when used to generate Arabic word embeddings. We first build a benchmark for the Arabic language that can be utilized to perform intrinsic evaluation of different word embeddings. We then perform additional extrinsic evaluations of the embeddings based on two NLP tasks. | Methodical Evaluation of Arabic Word Embeddings |
d248299679 | We propose DiffCSE, an unsupervised contrastive learning framework for learning sentence embeddings. DiffCSE learns sentence embeddings that are sensitive to the difference between the original sentence and an edited sentence, where the edited sentence is obtained by stochastically masking out the original sentence and then sampling from a masked language model. We show that DiffSCE is an instance of equivariant contrastive learning (Dangovski et al., 2021), which generalizes contrastive learning and learns representations that are insensitive to certain types of augmentations and sensitive to other "harmful" types of augmentations. Our experiments show that DiffCSE achieves state-of-the-art results among unsupervised sentence representation learning methods, outperforming unsupervised SimCSE 1 by 2.3 absolute points on semantic textual similarity tasks. 2 | DiffCSE: Difference-based Contrastive Learning for Sentence Embeddings |
d247597105 | We investigate what kind of structural knowledge learned in neural network encoders is transferable to processing natural language. We design artificial languages with structural properties that mimic natural language, pretrain encoders on the data, and see how much performance the encoder exhibits on downstream tasks in natural language. Our experimental results show that pretraining with an artificial language with a nesting dependency structure provides some knowledge transferable to natural language. A follow-up probing analysis indicates that its success in the transfer is related to the amount of encoded contextual information and what is transferred is the knowledge of position-aware context dependence of language. Our results provide insights into how neural network encoders process human languages and the source of cross-lingual transferability of recent multilingual language models. | Pretraining with Artificial Language: Studying Transferable Knowledge in Language Models |
d6068743 | This paper starts by giving a short overview of one existing NLP project for the medical sublanguage (1). After having presented our objectives (2), we will describe the Restriction Grammar formalism (3), the datastructure we use for parsing (4) and our parser(5)enhanced with a special control structure (6). An attempt to build a bootstrap dictionary of medical terminology in a semi-automatic way will also be discussed(7). A brief evaluation (8) and a short outline of our future research (9) will conclude this article.1. context and state of the art ACTES DE COLING-92, NANTES, 23-28 Adler 1992 Zweigenbaum P. et al (1990): Natural language processing of patient discharge summaries (NLPAD)extraction prototype, in Noothoven van Goor | APPLYING AND IMPROVING THE RESTRICTION GRAMMAR APPROACH FOR DUTCH PATIENT DISCHARGE SUMMARIES |
d258740707 | An important problem of the sequence-tosequence neural models widely used in abstractive summarization is exposure bias. To alleviate this problem, re-ranking systems have been applied in recent years. Despite some performance improvements, this approach remains underexplored. Previous works have mostly specified the rank through the ROUGE score and aligned candidate summaries, but there can be quite a large gap between the lexical overlap metric and semantic similarity. In this paper, we propose a novel training method in which a re-ranker balances the lexical and semantic quality. We further newly define false positives in ranking and present a strategy to reduce their influence. Experiments on the CNN/DailyMail and XSum datasets show that our method can estimate the meaning of summaries without seriously degrading the lexical aspect. More specifically, it achieves an 89.67 BERTScore on the CNN/DailyMail dataset, reaching new state-of-the-art performance. Our code is publicly available at https: //github.com/jeewoo1025/BalSum. | Balancing Lexical and Semantic Quality in Abstractive Summarization |
d8767164 | GENERATING ENGLISH PARAPHRASES FROM FORMAL RELATIONAL CALCUq.US EX~RESSIONS | |
d252816067 | Billions of people across the globe have been using social media platforms in their local languages to voice their opinions about the various topics related to the COVID-19 pandemic. Several organizations, including the World Health Organization, have developed automated social media analysis tools that classify COVID-19-related tweets to various topics. However, these tools that help combat the pandemic are limited to very few languages, making several countries unable to take their benefit. While multi-lingual or low-resource languagespecific tools are being developed, there is still a need to expand their coverage, such as for the Nepali language. In this paper, we identify the eight most common COVID-19 discussion topics among the Twitter community using the Nepali language, set up an online platform to automatically gather Nepali tweets containing the COVID-19-related keywords, classify the tweets into the eight topics, and visualize the results across the period in a web-based dashboard. We compare the performance of two state-of-the-art multi-lingual language models for Nepali tweet classification, one generic (mBERT) and the other Nepali language familyspecific model (MuRIL). Our results show that the models' relative performance depends on the data size, with MuRIL doing better for a larger dataset. The annotated data, models, and the web-based dashboard are open-sourced at https://github.com/naamiinepal/cov id-tweet-classification. | COVID-19-related Nepali Tweets Classification in a Low Resource Setting |
d252818997 | Event Causality Identification (ECI), which aims to detect whether a causality relation exists between two given textual events, is an important task for event causality understanding. However, the ECI task ignores crucial event structure and cause-effect causality component information, making it struggle for downstream applications. In this paper, we explore a novel task, namely Event Causality Extraction (ECE), aiming to extract the cause-effect event causality pairs with their structured event information from plain texts. The ECE task is more challenging since each event can contain multiple event arguments, posing fine-grained correlations between events to decide the causeeffect event pair. Hence, we propose a method with a dual grid tagging scheme to capture the intra-and inter-event argument correlations for ECE. Further, we devise a event type-enhanced model architecture to realize the dual grid tagging scheme. Experiments demonstrate the effectiveness of our method, and extensive analyses point out several future directions for ECE. * Corresponding AuthorThe worldwide rise of oil prices increases the cost of international shipping industry and stimulates the demand for new energy such as Ammonia fuel. | Event Causality Extraction with Event Argument Correlations |
d254926860 | Two key obstacles in biomedical relation extraction (RE) are the scarcity of annotations and the prevalence of instances without explicitly pre-defined labels due to low annotation coverage. Existing approaches, which treat biomedical RE as a multi-class classification task, often result in poor generalization in low-resource settings and do not have the ability to make selective predictions on unknown cases but give a guess from seen relations, hindering the applicability of those approaches. We present NBR, which converts biomedical RE as a natural language inference formulation to provide indirect supervision. By converting relations to natural language hypotheses, NBR is capable of exploiting semantic cues to alleviate annotation scarcity. By incorporating a ranking-based loss that implicitly calibrates abstinent instances, NBR learns a clearer decision boundary and is instructed to abstain on uncertain instances. Extensive experiments on three widely-used biomedical RE benchmarks, namely ChemProt, DDI, and GAD, verify the effectiveness of NBR in both full-shot and low-resource regimes. Our analysis demonstrates that indirect supervision benefits biomedical RE even when a domain gap exists, and combining NLI knowledge with biomedical knowledge leads to the best performance gains. 1 | Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction? |
d237366367 | The automatic evaluation of machine translation plays an important role in promoting the development and application of machine translation. It generally measures the quality of machine translation through calculating the similarity between machine translation and its reference. This paper uses the cross-lingual language model XLM to map source sentences, machine translations and reference to the same semantic space, and combines layer-wise attention and intra attention to extract the difference features from source sentences and machine translations, machine translations and its references, source sentences and its references, then integrates them into the automatic evaluation method based on neural network Bi-LSTM. The experimental results on the dataset of WMT'19 Metrics task show that the neural automatic evaluation method of machine translation combined with XLM word representation significantly improves its correlation with human judgments. | Neural Automatic Evaluation of Machine Translation Method Combined with XLM Word Representation |
d258461441 | Prior research has investigated the impact of various linguistic features on cross-lingual transfer performance. In this study, we investigate the manner in which this effect can be mapped onto the representation space. While past studies have focused on the impact on cross-lingual alignment in multilingual language models during fine-tuning, this study examines the absolute evolution of the respective language representation spaces produced by MLLMs. We place a specific emphasis on the role of linguistic characteristics and investigate their inter-correlation with the impact on representation spaces and cross-lingual transfer performance. Additionally, this paper provides preliminary evidence of how these findings can be leveraged to enhance transfer to linguistically distant languages. | Identifying the Correlation Between Language Distance and Cross-Lingual Transfer in a Multilingual Representation Space |
d236460087 | The cross-database context-dependent Text-to-SQL (XDTS) problem has attracted considerable attention in recent years due to its wide range of potential applications. However, we identify two biases in existing datasets for XDTS: (1) a high proportion of contextindependent questions and (2) a high proportion of easy SQL queries. These biases conceal the major challenges in XDTS to some extent. In this work, we present CHASE, a large-scale and pragmatic Chinese dataset for XDTS. It consists of 5,459 coherent question sequences (17,940 questions with their SQL queries annotated) over 280 databases, in which only 35% of questions are contextindependent, and 28% of SQL queries are easy. We experiment on CHASE with three state-ofthe-art XDTS approaches. The best approach only achieves an exact match accuracy of 40% over all questions and 16% over all question sequences, indicating that CHASE highlights the challenging problems of XDTS. We believe that CHASE can provide fertile soil for addressing the problems. * Work done during an internship at Microsoft Research. 哪所大学培养了最多 MVP 球员？ Which university has the most MVP players? SELECT T2.毕业院校 FROM MVP记录 AS T1 JOIN 球员 AS T2 ON T1.球员id = T2.球员id GROUP BY T2.毕业院校 ORDER BY COUNT(DISTINCT T2.球员id) DESC LIMIT 1; SELECT T2.college FROM MVP_record AS T1 JOIN player AS T2 ON T1.player_id = T2.player_id GROUP BY T2.college ORDER BY COUNT(DISTINCT T2.player_id) DESC LIMIT 1; 状元呢？ How about the first overall pick? 居然还是肯塔基！杜克也很出名啊，它培养了多少呢？ Still Kentucky! Duke is also very famous! How many does it have? SELECT 毕业院校 FROM 球员 WHERE 是否状元 = "是" GROUP BY 毕业院校 ORDER BY COUNT(*) DESC LIMIT 1; SELECT college FROM player WHERE is_first_pick = "yes" GROUP BY college ORDER BY COUNT(*) DESC LIMIT 1; SELECT COUNT(*) FROM 球员 WHERE 是否状元 = "是" AND 毕业院校 LIKE "%杜克%";SELECT COUNT(*) FROM player WHERE is_first_pick = "yes" AND college LIKE "%duke%"; | CHASE: A Large-Scale and Pragmatic Chinese Dataset for Cross-Database Context-Dependent Text-to-SQL |
d234762987 | Relation classification aims to predict a relation between two entities in a sentence. The existing methods regard all relations as the candidate relations for the two entities. These methods neglect the restrictions on candidate relations by entity types, which leads to some inappropriate relations being candidate relations. In this paper, we propose a novel paradigm, RElation Classification with ENtity Type restriction (RECENT), which exploits entity types to restrict candidate relations. Specially, the mutual restrictions of relations and entity types are formalized and introduced into relation classification. Besides, the proposed paradigm, RECENT, is model-agnostic. Based on two representative models GCN and SpanBERT respectively, RECENT GCN and RECENT SpanBERT are trained in RE-CENT 1 . Experimental results on a standard dataset indicate that RECENT improves the performance of GCN and SpanBERT by 6.9 and 4.4 F1 points, respectively. Especially, RECENT SpanBERT achieves a new state-ofthe-art on TACRED. | Relation Classification with Entity Type Restriction |
d252624586 | In free word association tasks, human subjects are presented with a stimulus word and are then asked to name the first word (the response word) that comes up to their mind. Those associations, presumably learned on the basis of conceptual contiguity or similarity, have attracted for a long time the attention of researchers in linguistics and cognitive psychology, since they are considered as clues about the internal organization of the lexical knowledge in the semantic memory. Word associations data have also been used to assess the performance of Vector Space Models for English, but evaluations for other languages have been relatively rare so far. In this paper, we introduce word associations datasets for Italian, Spanish and Mandarin Chinese by extracting data from the Small World of Words project, and we propose two different tasks inspired by the previous literature. We tested both monolingual and crosslingual word embeddings on the new datasets, showing that they perform similarly in the evaluation tasks. | Evaluating Monolingual and Crosslingual Embeddings on Datasets of Word Association Norms |
d52009317 | In this paper, we describe an attempt towards the development of parallel corpora for English and Ethiopian Languages, such as Amharic, Tigrigna, Afan-Oromo, Wolaytta and Ge'ez. The corpora are used for conducting a bi-directional statistical machine translation experiments. The BLEU scores of the bi-directional Statistical Machine Translation (SMT) systems show a promising result. The morphological richness of the Ethiopian languages has a great impact on the performance of SMT specially when the targets are Ethiopian languages. Now we are working towards an optimal alignment for a bidirectional English-Ethiopian languages SMT. | Parallel Corpora for bi-lingual English-Ethiopian Languages Statistical Machine Translation |
d232021523 | ||
d248218724 | This paper describes our system, which placed third in the Multilingual Track (subtask 11), fourth in the Code-Mixed Track (subtask 12), and seventh in the Chinese Track (subtask 9) in the SemEval 2022 Task 11: MultiCoNER Multilingual Complex Named Entity Recognition. Our system's key contributions are as follows: 1) For multilingual NER tasks, we offer an unified framework with which one can easily execute single-language or multilingual NER tasks, 2) for low-resource code-mixed NER task, one can easily enhance his or her dataset through implementing several simple data augmentation methods and 3) for Chinese tasks, we propose a model that can capture Chinese lexical semantic, lexical border, and lexical graph structural information. Finally, our system achieves macro-f1 scores of 77.66, 84.35, and 74.00 on subtasks 11, 12, and 9, respectively, during the testing phase. | Qtrade AI at SemEval-2022 Task 11: An Unified Framework for Multilingual NER Task |
d219310324 | ||
d218973949 | ||
d563162 | Most of the proposals for semantics in the Tree Adjoining Grammar (TAG) framework suppose that the derivation tree serves as basis for semantics. However, in some cases the derivation tree does not provide the semantic links one needs. This paper concentrates on one of these cases, namely the analysis of quantifiers as adjuncts. The paper proposes to enrich the TAG derivation tree and use the resulting structure as basis for semantics. This allows to deal with quantifiers, even in PPs embedded into NPs, such that an adequate semantics with appropriate scope orders is obtained. The enriched derivation structure allows also to treat other cases that are problematic for the assumption that a TAG semantics can be based on the derivation tree. c ¢ 2002 Laura Kallmeyer. | Using an Enriched TAG Derivation Structure as Basis for Semantics |
d32925047 | The paper presents a method for WordNet supersense tagging of Sanskrit, an ancient Indian language with a corpus grown over four millenia. The proposed method merges lexical information from Sanskrit texts with lexicographic definitions from Sanskrit-English dictionaries, and compares the performance of two machine learning methods for this task. Evaluation concentrates on Vedic, the oldest layer of Sanskrit. This level of Sanskrit contains numerous rare words that are no longer used in the later language and whose word senses can, therefore, not be induced from their occurrences in other texts. The paper studies how to efficiently transfer knowledge from later forms of Sanskrit and from modern Western dictionaries for this special task of supersense disambiguation. | Coarse Semantic Classification of Rare Nouns Using Cross-Lingual Data and Recurrent Neural Networks |
d248780365 | Identifying offensive speech is an exciting and essential area of research, with ample traction in recent times. This paper presents our system submission to the subtask 1, focusing on using supervised approaches for extracting Offensive spans from code-mixed Tamil-English comments. To identify offensive spans, we developed the Bidirectional Long Short-Term Memory (BiLSTM) model with Glove Embedding. With this method, the developed system achieved an overall F1 of 0.1728. Additionally, for comments with less than 30 characters, the developed system shows an F1 of 0.3890, competitive with other submissions. | DLRG@TamilNLP-ACL2022: Offensive Span Identification in Tamil using BiLSTM-CRF approach |
d237592974 | Several NLP tasks need the effective representation of text documents. Arora et al., 2017 demonstrate that simple weighted averaging of word vectors frequently outperforms neural models. SCDV(Mekala et al., 2017)further extends this from sentences to documents by employing soft and sparse clustering over pre-computed word vectors. However, both techniques ignore the polysemy and contextual character of words. In this paper, we address this issue by proposing SCDV+BERT(ctxd), a simple and effective unsupervised representation that combines contextualized BERT (Devlin et al., 2019) based word embedding for word sense disambiguation with SCDV soft clustering approach. We show that our embeddings outperform original SCDV, pre-train BERT, and several other baselines on many classification datasets. We also demonstrate our embeddings effectiveness on other tasks, such as concept matching and sentence similarity. In addition, we show that SCDV+BERT(ctxd) outperforms fine-tune BERT and different embedding approaches in scenarios with limited data and only few shots examples. | Unsupervised Contextualized Document Representation |
d231709509 | Natural language processing (NLP) tasks (e.g. question-answering in English) benefit from knowledge of other tasks (e.g., named entity recognition in English) and knowledge of other languages (e.g., question-answering in Spanish). Such shared representations are typically learned in isolation, either across tasks or across languages. In this work, we propose a meta-learning approach to learn the interactions between both tasks and languages. We also investigate the role of different sampling strategies used during meta-learning. We present experiments on five different tasks and six different languages from the XTREME multilingual benchmark dataset (Hu et al., 2020). Our meta-learned model clearly improves in performance compared to competitive baseline models that also include multitask baselines. We also present zero-shot evaluations on unseen target languages to demonstrate the utility of our proposed model. | Meta-Learning for Effective Multi-task and Multilingual Modelling |
d225062870 | ||
d222133300 | ||
d16133005 | Punjabi and Hindi are two closely related languages as both originated from the same origin and having lot of syntactic and semantic similarities. These similarities make direct translation methodology an obvious choice for Punjabi-Hindi language pair. The purposed system for Punjabi to Hindi translation has been implemented with various research techniques based on Direct MT architecture and language corpus. The output is evaluated by already prescribed methods in order to get the suitability of the system for the Punjabi Hindi language pair. | Coling 2008: Companion volume -Posters and Demonstrations |
d16648836 | We present the first English syllabification system to improve the accuracy of letter-tophoneme conversion. We propose a novel discriminative approach to automatic syllabification based on structured SVMs. In comparison with a state-of-the-art syllabification system, we reduce the syllabification word error rate for English by 33%. Our approach also performs well on other languages, comparing favorably with published results on German and Dutch. | Automatic Syllabification with Structured SVMs for Letter-To-Phoneme Conversion |
d225062808 | Information extraction from documents such as receipts or invoices is a fundamental and crucial step for office automation. Many approaches focus on extracting entities and relationships from plain texts, however, when it comes to document images, such demand becomes quite challenging since visual and layout information are also of great significance to help tackle this problem. In this work, we propose the attention-based graph neural network to combine textual and visual information from document images. Moreover, the global node is introduced in our graph construction algorithm which is used as a virtual hub to collect the information from all the nodes and edges to help improve the performance. Extensive experiments on real-world datasets show that our method outperforms baseline methods by significant margins. | Attention-Based Graph Neural Network with Global Context Awareness for Document Understanding |
d16125805 | Human-targeted metrics provide a compromise between human evaluation of machine translation, where high inter-annotator agreement is difficult to achieve, and fully automatic metrics, such as BLEU or TER, that lack the validity of human assessment. Human-targeted translation edit rate (HTER) is by far the most widely employed human-targeted metric in machine translation, commonly employed, for example, as a gold standard in evaluation of quality estimation. Original experiments justifying the design of HTER, as opposed to other possible formulations, were limited to a small sample of translations and a single language pair, however, and this motivates our re-evaluation of a range of human-targeted metrics on a substantially larger scale. Results show significantly stronger correlation with human judgment for HBLEU over HTER for two of the nine language pairs we include and no significant difference between correlations achieved by HTER and HBLEU for the remaining language pairs. Finally, we evaluate a range of quality estimation systems employing HTER and direct assessment (DA) of translation adequacy as gold labels, resulting in a divergence in system rankings, and propose employment of DA for future quality estimation evaluations. | Is all that Glitters in Machine Translation Quality Estimation really Gold? |
d229365842 | ||
d252407528 | Incorporating personal preference is crucial in advanced machine translation tasks. Despite the recent advancement of machine translation, it remains a demanding task to properly reflect personal style. In this paper, we introduce a personalized automatic post-editing framework to address this challenge, which effectively generates sentences considering distinct personal behaviors. To build this framework, we first collect post-editing data that connotes the user preference from a live machine translation system. Specifically, real-world users enter source sentences for translation and edit the machinetranslated outputs according to the user's preferred style. We then propose a model that combines a discriminator module and user-specific parameters on the APE framework. Experimental results show that the proposed method outperforms other baseline models on four different metrics (i.e., BLEU, TER, YiSi-1, and human evaluation). | PePe: Personalized Post-editing Model utilizing User-generated Post-edits |
d254564585 | Recent pre-trained language models have shown promising capabilities in generating fluent and realistic natural language text. However, generating multi-sentence text with global content planning has been a long-existing research question. Current approaches for controlled text generation can hardly address this issue, as they usually condition on single known control attributes. In this study, we propose a low-cost yet effective framework which explicitly models the global content plan of the generated text. Specifically, it optimizes the joint distribution of the natural language sequence and the global content plan in a plugand-play manner. We conduct extensive experiments on the well-established Recipe1M+ benchmark. Both automatic and human evaluations verify that our model achieves the stateof-the-art performance on the task of recipe generation. 1 | Plug-and-Play Recipe Generation with Content Planning |
d232075938 | The increasing accessibility of the internet facilitated social media usage and encouraged individuals to express their opinions liberally. Nevertheless, it also creates a place for content polluters to disseminate offensive posts or contents. Most of such offensive posts are written in a cross-lingual manner and can easily evade the online surveillance systems. This paper presents an automated system that can identify offensive text from multilingual code-mixed data.In the task, datasets provided in three languages including Tamil, Malayalam and Kannada code-mixed with English where participants are asked to implement separate models for each language.To accomplish the tasks, we employed two machine learning techniques (LR, SVM), three deep learning (LSTM, LSTM+Attention) techniques and three transformers (m-BERT, Indic-BERT, XLM-R) based methods. Results show that XLM-R outperforms other techniques in Tamil and Malayalam languages while m-BERT achieves the highest score in the Kannada language.The proposed models gained weighted f 1 score of 0.76 (for Tamil), 0.93 (for Malayalam), and 0.71 (for Kannada) with a rank of 3 rd , 5 th and 4 th respectively. | NLP-CUET@DravidianLangTech-EACL2021: Offensive Language Detection from Multilingual Code-Mixed Text using Transformers |
d233365107 | This study sheds light on the effects of COVID-19 in the particular field of Computational Linguistics and Natural Language Processing within Artificial Intelligence. We provide an inter-sectional study on gender, contribution, and experience that considers one school year (from August 2019 to August 2020) as a pandemic year. August is included twice for the purpose of an inter-annual comparison.While there has been an increasing trend in publications during the crisis, the results show that the ratio between female authors' publications and male authors' publications decreased. This only reduces the importance of the female role in the scientific contributions of computational linguistics (it is now far below its peak of 0.24). The pandemic has a significantly negative effect on female senior researchers' production in the first position of authors (maximum work), followed by the female junior researchers in the last position of authors (supervision or collaborative work). | |
d234762800 | Given a video, video grounding aims to retrieve a temporal moment that semantically corresponds to a language query. In this work, we propose a Parallel Attention Network with Sequence matching (SeqPAN) to address the challenges in this task: multi-modal representation learning, and target moment boundary prediction. We design a self-guided parallel attention module to effectively capture selfmodal contexts and cross-modal attentive information between video and text. Inspired by sequence labeling tasks in natural language processing, we split the ground truth moment into begin, inside, and end regions. We then propose a sequence matching strategy to guide start/end boundary predictions using region labels. Experimental results on three datasets show that SeqPAN is superior to state-of-theart methods. Furthermore, the effectiveness of the self-guided parallel attention module and the sequence matching module is verified. 1 | Parallel Attention Network with Sequence Matching for Video Grounding |
d227231713 | ||
d236460081 | Recent years have brought about an interest in the challenging task of summarizing conversation threads (meetings, online discussions, etc.). Such summaries help analysis of the long text to quickly catch up with the decisions made and thus improve our work or communication efficiency. To spur research in thread summarization, we have developed an abstractive Email Thread Summarization (EMAILSUM) dataset, which contains humanannotated short (<30 words) and long (<100 words) summaries of 2,549 email threads (each containing 3 to 10 emails) over a wide variety of topics. We perform a comprehensive empirical study to explore different summarization techniques (including extractive and abstractive methods, single-document and hierarchical models, as well as transfer and semisupervised learning) and conduct human evaluations on both short and long summary generation tasks. Our results reveal the key challenges of current abstractive summarization models in this task, such as understanding the sender's intent and identifying the roles of sender and receiver. Furthermore, we find that widely used automatic evaluation metrics (ROUGE, BERTScore) are weakly correlated with human judgments on this email thread summarization task. Hence, we emphasize the importance of human evaluation and the development of better metrics by the community. 1 | EMAILSUM: Abstractive Email Thread Summarization |
d6605351 | The task of documenting the world's languages is a mainstream activity in linguistics which is yet to spill over into computational linguistics. We propose a new task of transcription normalisation as an algorithmic method for speeding up the process of transcribing audio sources, leading to text collections of usable quality. We report on the application of sentence and word alignment algorithms to this task, before describing a new algorithm. All of the algorithms are evaluated over synthetic datasets. Although the results are nuanced, the transcription normalisation task is suggested as an NLP contribution to the grand challenge of documenting the world's languages. | Normalising Audio Transcriptions for Unwritten Languages |
d394071 | This paper describes an approach to improve summaries for a collection of Twitter posts created using the Phrase Reinforcement (PR) Algorithm (Sharifi et al., 2010a). The PR algorithm often generates summaries with excess text and noisy speech. We parse these summaries using a dependency parser and use the dependencies to eliminate some of the excess text and build better-formed summaries. We compare the results to those obtained using the PR Algorithm. | Better Twitter Summaries? |
d256827670 | With the power of large pretrained language models, various research works have integrated knowledge into dialogue systems. The traditional techniques treat knowledge as part of the input sequence for the dialogue system, prepending a set of knowledge statements in front of dialogue history. However, such a mechanism forces knowledge sets to be concatenated in an ordered manner, making models implicitly pay imbalanced attention to the sets during training. In this paper, we first investigate how the order of the knowledge set can influence autoregressive dialogue systems' responses. We conduct experiments on two commonly used dialogue datasets with two types of transformer-based models and find that models view the input knowledge unequally. To this end, we propose a simple and novel technique to alleviate the order effect by modifying the position embeddings of knowledge input in these models. With the proposed position embedding method, the experimental results show that each knowledge statement is uniformly considered to generate responses. * Work done when interning at Intel Labs. | Position Matters! Empirical Study of Order Effect in Knowledge-grounded Dialogue |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.