L3Cube-MahaSTS: A Marathi Sentence Similarity Dataset and Models
Not on the map. Why?
Abstract (arXiv)
We present MahaSTS, a human-annotated Sentence Textual Similarity (STS) dataset for Marathi, along with MahaSBERT-STS-v2, a fine-tuned Sentence-BERT model optimized for regression-based similarity scoring. The MahaSTS dataset consists of 16,860 Marathi sentence pairs labeled with continuous similarity scores in the range of 0-5. To ensure balanced supervision, the dataset is uniformly distributed across six score-based buckets spanning the full 0-5 range, thus reducing label bias and enhancing model stability. We fine-tune the MahaSBERT model on this dataset and benchmark its performance against other alternatives like MahaBERT, MuRIL, IndicBERT, and IndicSBERT. Our experiments demonstrate that MahaSTS enables effective training for sentence similarity tasks in Marathi, highlighting the impact of human-curated annotations, targeted fine-tuning, and structured supervision in low-resource settings. The dataset and model are publicly shared at https://github.com/l3cube-pune/MarathiNLP
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 15 of its 18 references that are in the map. First posted on arXiv 29 Aug 2025.
Builds on
15 of 18 references.
References · 15
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
- 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
- 2017SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluationcited by 1,001
- 2020Language-agnostic BERT Sentence Embeddingcited by 627
- 2017Supervised Learning of Universal Sentence Representations from Natural Language Inference Datacited by 432
- 2022L3Cube-MahaCorpus and MahaBERT: Marathi Monolingual Corpus, Marathi BERT Language Models, and Resourcescited by 51
- 2020GottBERT: a pure German Language Modelcited by 49
- 2018Evaluating Compositionality in Sentence Embeddingscited by 35
- 2021RobeCzech: Czech RoBERTa, a monolingual contextualized language representation modelcited by 33
- 2022Mono vs Multilingual BERT for Hate Speech Detection and Text Classification: A Case Study in Marathicited by 16
3 more
- 2023L3Cube-IndicSBERT: A simple approach for learning cross-lingual sentence representations using multilingual BERTcited by 15
- 2022L3Cube-MahaSBERT and HindSBERT: Sentence BERT Models and Benchmarking BERT Sentence Representations for Hindi and Marathicited by 10
- 2018Improving Multilingual Semantic Textual Similarity with Shared Sentence Encoder for Low-resource Languagescited by 5
Built on by
Citing papers per year: 2026: 1.
1 paper.