Maithili is one of the 22 official languages recognized in the Indian Constitution. The literature of Maithili is rich; however, due to current socio-political changes, the language is on the verge of extinction. Therefore, it is crucial to develop a corpus for low-resource Indic languages like Maithili to ensure that the dream of “No Language Left Behind" (NLLB) is realized. With this in mind, we contribute a corpus (1,05,600 sentences) containing both manually curated and synthetically generated. Additionally, we propose a strong baseline on the Maithali-Hindi pair using multilingual pretrained models such as IndicTrans2, mBART50, mT5, and NLLB-200 distilled. We evaluate the translation systems using standard performance metrics, including BLEU, CHRF2, TER, COMET, METEOR, and BERTScore. Comparative experiments conducted against the existing NLLB dataset (5,50,300 sentence pairs) demonstrate that our proposed dataset consistently yields superior translation quality. Finally, these results demonstrate that, even with a smaller corpus size, high-quality, task-specific data significantly enhance translation accuracy for low-resource Indian languages, such as Maithili.

This paper reports on the outcomes of the shared tasks organized as part of the 23rd International Workshop on Spoken Language Translation (IWSLT). The workshop covered ten major challenges in spoken language translation, including speech-to-text translation for both high-resource and low-resource language pairs, customized speech translation, speech generation, instruction-following speech processing, and the evaluation of speech translation systems. The shared tasks received strong participation, with more than 30 teams submitting runs. This year’s edition broadened the range of tasks, placing particular emphasis on speech generation and evaluation metrics.

Cross-lingual learning enables the transfer of structured sentiment knowledge from high-resource languages to unlabeled or low-resource languages, but prior work has largely focused on coarse-grained sentiment classification or aspect extraction. In contrast, zero-shot cross-lingual aspect–opinion–sentiment triplet extraction (ASTE), which extracts sentiment triplets of the form (aspect term, opinion term, sentiment polarity), remains underexplored. We propose a unified framework that leverages large language models (LLMs) as both structured pseudo-label generators and semantic teachers for ASTE. Our approach employs stepwise structured prompting over aspect- and opinion-aware code-switched variants to generate reliable pseudo triplets, followed by a multi-variant consistency filter to retain high-confidence supervision. We further introduce a triplet-aware contrastive distillation objective that aligns student triplet representations with LLM-encoded semantic embeddings. During inference, only the student ASTE model is used, without requiring LLM access. Experiments on four non-Indic and four low-resource Indic target languages show consistent improvements over strong cross-lingual and LLM-based baselines. The proposed method yields an absolute micro-F1 improvement of 5.3 points on non-Indic languages and 3.8 points on low-resource Indic languages compared to the best competing approach. Ablation results further validate the complementary roles of aspect- and opinion-aware code-switched prompting and triplet-aware contrastive distillation, with larger relative gains observed in low-resource Indic settings.

This paper presents the outcomes of the shared tasks conducted at the 22nd International Workshop on Spoken Language Translation (IWSLT). The workshop addressed seven critical challenges in spoken language translation: simultaneous and offline translation, automatic subtitling and dubbing, model compression, speech-to-speech translation, dialect and low-resource speech translation, and Indic languages. The shared tasks garnered significant participation, with 32 teams submitting their runs. The field’s growing importance is reflected in the increasing diversity of shared task organizers and contributors to this overview paper, representing a balanced mix of industrial and academic institutions. This broad participation demonstrates the rising prominence of spoken language translation in both research and practical applications.

Speech-to-Speech Translation (S2ST) converts speech from one language to speech in a different language. While various S2ST models exist, none adequately support Indic languages, primarily due to the lack of a suitable dataset. We fill this gap by introducing Indic-S2ST, a multilingual and multimodal many-to-many S2ST data of approximately 600 hours in 14 Indic languages, including Indian-accented English. To the best of our knowledge, this is the largest data for the S2ST task with parallel speech and text in 14 scheduled Indic languages. Our data also supports Automatic Speech Recognition (ASR), Text-to-Speech (TTS) synthesis, Speech-to-Text translation (ST), and Machine Translation (MT) due to parallel speech and text alignment. Thus, our data may be useful to train a model likeMeta’s SeamlessM4T for Indic languages. We also propose Indic-S2UT, a discrete unit-based S2ST model for Indic languages. To showcase the utility of the data, we present baseline results on the Indic-S2ST data using the Indic-S2UT. The dataset and codes are available at https://github.com/Nivedita5/Indic-S2ST/blob/main/README.md.

The rapid growth of online product reviews spurs significant interest in Aspect-Based Sentiment Analysis (ABSA), which involves identifying aspect terms and their associated sentiment polarity. While ABSA is widely studied in resource-rich languages like English, Chinese, and Spanish, it remains underexplored in low-resource languages such as Odia. To address this gap, we create a reliable resource for aspect-based sentiment analysis in Odia. The dataset is annotated for two specific tasks: Aspect Term Extraction (ATE) and Aspect Polarity Classification (APC), spanning seven domains and aligned with the SemEval-2014 benchmark. Furthermore, we employ an ensemble data augmentation approach combining back-translation with a fine-tuned T5 paraphrase generation model to enhance the dataset and apply a semantic similarity filter using a Universal Sentence Encoder (USE) to remove low-quality data and ensure a balanced distribution of sample difficulty in the newly augmented dataset. Finally, we validate our dataset by fine-tuning multilingual pre-trained models, XLM-R and IndicBERT, on ATE and APC tasks. Additionally, we use three classical baseline models to evaluate the quality of the proposed dataset for these tasks. We hope the Odia dataset will spur more work for the ABSA task.

Speech-to-text (ST) task is the translation of speech in a language to text in a different language. It has use cases in subtitling, dubbing, etc. Traditionally, ST task has been solved by cascading automatic speech recognition (ASR) and machine translation (MT) models which leads to error propagation, high latency, and training time. To minimize such issues, end-to-end models have been proposed recently. However, we find that only a few works have reported results of ST models on a limited number of low-resource languages. To take a step further in this direction, we release datasets and baselines for low-resource ST tasks. Concretely, our dataset has 9 language pairs and benchmarking has been done against SOTA ST models. The low performance of SOTA ST models on Indic-TEDST data indicates the necessity of the development of ST models specifically designed for low-resource languages.

This paper reports on the shared tasks organized by the 21st IWSLT Conference. The shared tasks address 7 scientific challenges in spoken language translation: simultaneous and offline translation, automatic subtitling and dubbing, speech-to-speech translation, dialect and low-resource speech translation, and Indic languages. The shared tasks attracted 17 teams whose submissions are documented in 27 system papers. The growing interest towards spoken language translation is also witnessed by the constantly increasing number of shared task organizers and contributors to the overview paper, almost evenly distributed across industry and academia.

The presence of sarcasm in conversational systems and social media like chatbots, Facebook, Twitter, etc. poses several challenges for downstream NLP tasks. This is attributed to the fact that the intended meaning of a sarcastic text is contrary to what is expressed. Further, the use of code-mix language to express sarcasm is increasing day by day. Current NLP techniques for code-mix data have limited success due to the use of different lexicon, syntax, and scarcity of labeled corpora. To solve the joint problem of code-mixing and sarcasm detection, we propose the idea of capturing incongruity through sub-word level embeddings learned via fastText. Empirical results show that our proposed model achieves an F1-score on code-mix Hinglish dataset comparable to pretrained multilingual models while training 10x faster and using a lower memory footprint.