Abstract

Evaluating machine translation (MT) systems for low-resource languages has long been a challenge due to the limited availability of evaluation metrics and resources. As a result, researchers in this space have relied primarily on lexical-based metrics like BLEU, TER, and ChrF, which lack semantic evaluation. In this first-of-its-kind work, we propose a novel pivot-based evaluation framework that addresses these limitations; after translating low-resource language outputs into a related high-resource language, we leverage advanced neural and embedding-based metrics for more meaningful evaluation. Through a series of experiments using five low-resource languages: Assamese, Manipuri, Kannada, Bhojpuri, and Nepali, we demonstrate how this method extends the coverage of both lexical-based and embedding-based metrics, even for languages not directly supported by advanced metrics. Our results show that the differences between direct and pivot-based evaluation scores are minimal, proving that this approach is a viable and effective solution for evaluating translations in endangered and low-resource languages. This work paves the way for more inclusive, accurate, and scalable MT evaluation for underrepresented languages, marking a significant step forward in this under-explored area of research.

Abstract

This paper presents an evaluation of 16 machine translation systems submitted to the Shared Task of the 9th Conference of Machine Translation (WMT24) for the English-Hindi (en-hi) language pair using our Complex Structures Test (CoST) suite. Aligning with this year’s test suite sub-task theme, “Help us break LLMs”, we curated a comprehensive test suite encompassing diverse datasets across various categories, including autobiography, poetry, legal, conversation, play, narration, technical, and mixed genres. Our evaluation reveals that all the systems struggle significantly with the archaic style of text like legal and technical writings or text with creative twist like conversation and poetry datasets, highlighting their weaknesses in handling complex linguistic structures and stylistic nuances inherent in these text types. Our evaluation identifies the strengths and limitations of the submitted models, pointing to specific areas where further research and development are needed to enhance their performance. Our test suite is available at https://github.com/AnanyaCoder/CoST-WMT-24-Test-Suite-Task.

Abstract

Machine translation (MT) evaluation metrics like BLEU and chrF++ are widely used reference-based metrics that do not require training and are language-independent. However, these metrics primarily focus on n-gram matching and often overlook semantic depth and contextual understanding. To address this gap, we introduce chrF-S (Semantic chrF++), an enhanced metric that integrates sentence embeddings to evaluate translation quality more comprehensively. By combining traditional character and word n-gram analysis with semantic information derived from embeddings, chrF-S captures both syntactic accuracy and sentence-level semantics. This paper presents our contributions to the WMT24 shared metrics task, showcasing our participation and the development of chrF-S. We also demonstrate that, according to preliminary results on the leaderboard, our metric performs on par with other supervised and LLM-based metrics. By merging semantic insights with n-gram precision, chrF-S offers a significant enhancement in the assessment of machine-generated translations, advancing the field of MT evaluation. Our code and data will be made available at https://github.com/AnanyaCoder/chrF-S.

Abstract

Translating for languages with limited resources poses a persistent challenge due to the scarcity of high-quality training data. To enhance translation accuracy, we explored controlled generation mechanisms, focusing on the importance of control tokens. In our experiments, while training, we encoded the target sentence length as a control token to the source sentence, treating it as an additional feature for the source sentence. We developed various NMT models using transformer architecture and conducted experiments across 8 language directions (English = Assamese, Manipuri, Khasi, and Mizo), exploring four variations of length encoding mechanisms. Through comparative analysis against the baseline model, we submitted two systems for each language direction. We report our findings for the same in this work.

Abstract

Existing datasets for tabular question answering typically focus exclusively on text within cells. However, real-world data is inherently multimodal, often blending images such as symbols, faces, icons, patterns, and charts with textual content in tables. With the evolution of AI models capable of multimodal reasoning, it is pertinent to assess their efficacy in handling such structured data. This study investigates whether current AI models can perform knowledge-aware reasoning on multimodal structured data. We explore their ability to reason on tables that integrate both images and text, introducing MMTabQA, a new dataset designed for this purpose. Our experiments highlight substantial challenges for current AI models in effectively integrating and interpreting multiple text and image inputs, understanding visual context, and comparing visual content across images. These findings establish our dataset as a robust benchmark for advancing AI's comprehension and capabilities in analyzing multimodal structured data.

Abstract

Exploring and quantifying semantic relatedness is central to representing language and holds significant implications across various NLP tasks. While earlier NLP research primarily focused on semantic similarity, often within the English language context, we instead investigate the broader phenomenon of semantic relatedness. In this paper, we present SemRel, a new semantic relatedness dataset collection annotated by native speakers across 13 languages: Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Spanish, and Telugu. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia – regions characterised by a relatively limited availability of NLP resources. Each instance in the SemRel datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. The scores are obtained using a comparative annotation framework. We describe the data collection and annotation processes, challenges when building the datasets, baseline experiments, and their impact and utility in NLP.

Abstract

We present the first shared task on Semantic Textual Relatedness (STR). While earlier shared tasks primarily focused on semantic similarity, we instead investigate the broader phenomenon of semantic relatedness across 14 languages: Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Punjabi, Spanish, and Telugu. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia – regions characterised by the relatively limited availability of NLP resources. Each instance in the datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. Participating systems were asked to rank sentence pairs by their closeness in meaning (i.e., their degree of semantic relatedness) in the 14 languages in three main tracks: (a) supervised, (b) unsupervised, and (c) crosslingual. The task attracted 163 participants. We received 70 submissions in total (across all tasks) from 51 different teams, and 38 system description papers. We report on the best-performing systems as well as the most common and the most effective approaches for the three different tracks.

Abstract

In recent studies, the extensive utilization of large language models has underscored the im- portance of robust evaluation methodologies for assessing text generation quality and rel- evance to specific tasks. This has revealed a prevalent issue known as hallucination, an emergent condition in the model where gener- ated text lacks faithfulness to the source and deviates from the evaluation criteria. In this study, we formally define hallucination and pro- pose a framework for its quantitative detection in a zero-shot setting, leveraging our definition and the assumption that model outputs entail task and sample specific inputs. In detecting hallucinations, our solution achieves an accu- racy of 0.78 in a model-aware setting and 0.61 in a model-agnostic setting. Notably, our so- lution maintains computational efficiency, re- quiring far less computational resources than other SOTA approaches, aligning with the trend towards lightweight and compressed models.

Abstract

Conversation is the most natural form of human communication, where each utterance can range over a variety of possible emotions. While significant work has been done towards the detection of emotions in text, relatively little work has been done towards finding the cause of the said emotions, especially in multimodal settings. SemEval 2024 introduces the task of Multimodal Emotion Cause Analysis in Conversations, which aims to extract emotions reflected in individual utterances in a conversation involving multiple modalities (textual, audio, and visual modalities) along with the corresponding utterances that were the cause for the emotion. In this paper, we propose models that tackle this task as an utterance labeling and a sequence labeling problem and perform a comparative study of these models, involving baselines using different encoders, using BiLSTM for adding contextual information of the conversation, and finally adding a CRF layer to try to model the inter-dependencies between adjacent utterances more effectively. In the official leaderboard for the task, our architecture was ranked 8th, achieving an F1-score of 0.1759 on the leaderboard.

Abstract

While significant work has been done in the field of NLP on vertical thinking, which involves primarily logical thinking, little work has been done towards lateral thinking, which involves looking at problems from an unconventional perspective and defying existing conceptions and notions. Towards this direction, SemEval 2024 introduces the task of BRAINTEASER, which involves two types of questions -- Sentence Puzzles and Word Puzzles that defy conventional common-sense reasoning and constraints. In this paper, we tackle both types of questions using few-shot prompting on GPT-3.5 and gain insights regarding the difference in the nature of the two types. Our prompting strategy placed us 26th on the leaderboard for the Sentence Puzzle and 15th on the Word Puzzle task.