Articles

AbstractOnline articles using striking headlines that promise intriguing information are often used to attract readers. Most of the time, the information provided in the text is disappointing to the reader after the headline promised exciting news. As part of the SemEval-2023 challenge, we propose a system to generate a spoiler …...

AbstractIn the article, we present the CodeNLP submission to the SemEval-2023 Task 2: MultiCoNER II Multilingual Complex Named Entity Recognition. Our approach is based on data augmentation by combining various strategies of sequence generation for training. We show that the extended procedure of fine-tuning a pre-trained language model can bring …...

AbstractIn this work, we present a Named Entity Recognition (NER) system that was trained using a Frustratingly Easy Domain Adaptation (FEDA) over multiple legal corpora. The goal was to create a NER capable of detecting 14 types of legal named entities in Indian judgments. Besides the FEDA architecture, we explored …...

AbstractThis paper describes the HEVS-TUW team submission to the SemEval-2023 Task 8: Causal Claims. We participated in two subtasks: (1) causal claims detection and (2) PIO identification. For subtask 1, we experimented with an ensemble of weakly supervised question detection and fine-tuned Transformer-based models. For subtask 2 of PIO frame …...

AbstractThis paper describes the performance of a system which uses stance as an output instead of taking it as an input to identify 20 human values behind given arguments, based on two datasets for SemEval-2023 Task 4. The rationale was to draw a conclusion on whether predicting stance would help …...

AbstractThis paper describes our system submitted to SemEval-2023 Task 5: Clickbait Spoiling. We work on spoiler generation of the subtask 2 and develop a system which comprises two parts: 1) simple seq2seq spoiler generation and 2) post-hoc model ensembling. Using this simple method, we address the challenge of generating multipart …...

AbstractVisual Word Sense Disambiguation (VWSD) task aims to find the most related image among 10 images to an ambiguous word in some limited textual context. In this work, we use AltCLIP features and a 3-layer standard transformer encoder to compare the cosine similarity between the given phrase and different images. …...

AbstractThis paper presents system descriptions of our submitted outputs for WebNLG Challenge 2023. We use mT5 in multi-task and multilingual settings to generate more fluent and reliable verbalizations of the given RDF triples. Furthermore, we introduce a partial decoding technique to produce more elaborate yet simplified outputs. Additionally, we demonstrate …...

AbstractWe present our solution for the Russian RDF002 to-text generation task of the WebNLG Challenge 2023. We use the pretrained large language model named FRED-T5 (Zmitrovich et al., 2023) to finetune on the train dataset. Also, we propose several types of prompt and run experiments to analyze their effectiveness. Our …...

AbstractThe WebNLG task consists of mapping a knowledge graph to a text verbalising the con- tent of that graph. The 2017 WebNLG edi- tion required participating systems to gener- ate English text from a set of DBpedia triples, while the 2020 WebNLG+ challenge addition- ally included generation into Russian and …...

AbstractLanguage-and-vision models have shown good performance in tasks such as image-caption matching and caption generation. However, it is challenging for such models to generate pragmatically correct captions, which adequately reflect what is happening in one image or several images. It is crucial to evaluate this behaviour to understand underlying reasons …...

AbstractIn any system that uses structured knowledge graph (KG) data as its underlying knowledge representation, KG-to-text generation is a useful tool for turning parts of the graph data into text that can be understood by humans. Recent work has shown that models that make use of pretraining on large amounts …...

AbstractIn Referring Expression Generation, model inputs are often composed of different representations, including the visual properties of the intended referent, its relative position and size, and the visual context. Yet, the extent to which this information influences the generation process of black-box neural models is largely unclear. We investigate the …...

AbstractThe task of Visual Question Generation (VQG) is to generate human-like questions relevant to the given image. As VQG is an emerging research field, existing works tend to focus only on resource-rich language such as English due to the availability of datasets. In this paper, we propose the first Bengali …...

AbstractWhile the fluency and coherence of Large Language Models (LLMs) in text generation have seen significant improvements, their competency in generating appropriate expressions of uncertainty remains limited.Using a multilingual closed-book QA task and GPT-3.5, we explore how well LLMs are calibrated and express certainty across a diverse set of languages, …...

AbstractMonitoring online customer reviews is important for business organizations to measure customer satisfaction and better manage their reputations. In this paper, we propose a novel dynamic Brand-Topic Model (dBTM) which is able to automatically detect and track brand-associated sentiment scores and polarity-bearing topics from product reviews organized in temporally ordered …...

AbstractProbing has become a go-to methodology for interpreting and analyzing deep neural models in natural language processing. However, there is still a lack of understanding of the limitations and weaknesses of various types of probes. In this work, we suggest a strategy for input-level intervention on naturalistic sentences. Using our …...

AbstractIn an information-seeking conversation, a user may ask questions that are under-specified or unanswerable. An ideal agent would interact by initiating different response types according to the available knowledge sources. However, most current studies either fail to or artificially incorporate such agent-side initiative. This work presents InSCIt, a dataset for …...

AbstractSemantic parsing maps natural language questions into logical forms, which can be executed against a knowledge base for answers. In real-world applications, the performance of a parser is often limited by the lack of training data. To facilitate zero-shot learning, data synthesis has been widely studied to automatically generate paired …...

AbstractPre-trained language models have been successful in many knowledge-intensive NLP tasks. However, recent work has shown that models such as BERT are not “structurally ready” to aggregate textual information into a [CLS] vector for dense passage retrieval (DPR). This “lack of readiness” results from the gap between language model pre-training …...