Lukas Stankevičius; Mantas Lukoševičius
Extracting Sentence Embeddings from Pretrained Transformer Models Journal Article
In: Applied Sciences, vol. 14, no. 19, pp. 8887, 2024, ISSN: 2076-3417.
Links | BibTeX | Tags: Deep learning, Natural language processing, Transformers
@article{Stankevi\v{c}iusLuko\v{s}evi\v{c}ius2024,
title = {Extracting Sentence Embeddings from Pretrained Transformer Models},
author = { Lukas Stankevi\v{c}ius and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=Extracting_sentence_embeddings_from_pretrained_transformer_models.pdf
https://www.mdpi.com/2076-3417/14/19/8887
https://arxiv.org/abs/2408.08073
https://arxiv.org/pdf/2408.08073},
doi = {10.3390/app14198887},
issn = {2076-3417},
year = {2024},
date = {2024-10-02},
urldate = {2024-10-02},
journal = {Applied Sciences},
volume = {14},
number = {19},
pages = {8887},
keywords = {Deep learning, Natural language processing, Transformers},
pubstate = {published},
tppubtype = {article}
}
Domas Grigaliūnas; Mantas Lukoševičius
Inference acceleration for large language models using "stairs" assisted greedy generation Proceedings Article
In: CEUR workshop proceedings of the 29th international conference on information society and university studies (IVUS 2024) Kaunas, Lithuania, May 17, 2024, pp. 165–174, CEUR-WS 2024.
Links | BibTeX | Tags: Deep learning, Natural language processing, Transformers
@inproceedings{Grigaliunas2024,
title = {Inference acceleration for large language models using "stairs" assisted greedy generation},
author = {Domas Grigali\={u}nas and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=LLMs_acceleration_using_stairs_assisted_generation.pdf
https://ceur-ws.org/Vol-3885/paper18.pdf
https://arxiv.org/abs/2407.19947
https://arxiv.org/pdf/2407.19947
https://www.linkedin.com/posts/lukosevicius_generativeai-opensourceai-llm-activity-7224413004539260930-3l3n},
doi = {10.48550/arXiv.2407.19947},
year = {2024},
date = {2024-07-29},
urldate = {2024-07-29},
booktitle = {CEUR workshop proceedings of the 29th international conference on information society and university studies (IVUS 2024) Kaunas, Lithuania, May 17, 2024},
volume = {3885},
pages = {165--174},
organization = {CEUR-WS},
howpublished = {arXiv},
keywords = {Deep learning, Natural language processing, Transformers},
pubstate = {published},
tppubtype = {inproceedings}
}
- https://mantas.info/get-publication/?f=LLMs_acceleration_using_stairs_assisted_g[...]
- https://ceur-ws.org/Vol-3885/paper18.pdf
- https://arxiv.org/abs/2407.19947
- https://arxiv.org/pdf/2407.19947
- https://www.linkedin.com/posts/lukosevicius_generativeai-opensourceai-llm-activi[...]
- doi:10.48550/arXiv.2407.19947
Brigita Vileikytė; Mantas Lukoševičius; Lukas Stankevičius
Sentiment Analysis of Lithuanian Online Reviews Using Large Language Models Proceedings Article
In: CEUR workshop proceedings of the 29th international conference on information society and university studies (IVUS 2024) Kaunas, Lithuania, May 17, 2024, pp. 249–258, CEUR-WS 2024.
Links | BibTeX | Tags: Deep learning, Natural language processing, Source code available, Transformers
@inproceedings{Vileikyte2024,
title = {Sentiment Analysis of Lithuanian Online Reviews Using Large Language Models},
author = {Brigita Vileikyt\.{e} and Mantas Luko\v{s}evi\v{c}ius and Lukas Stankevi\v{c}ius},
url = {https://mantas.info/get-publication/?f=Lithuanian_sentiment_analysis_using_LLMs.pdf
https://ceur-ws.org/Vol-3885/paper27.pdf
https://arxiv.org/abs/2407.19914
https://arxiv.org/pdf/2407.19914
https://huggingface.co/brivil1/lithuanian-sentiment-analysis-DistilBERT
https://huggingface.co/brivil1/lithuanian-sentiment-analysis-ByT5
https://www.linkedin.com/posts/lukosevicius_sentiment-analysis-of-lithuanian-online-reviews-activity-7234181985559035906-lUi1},
doi = {10.48550/arXiv.2407.19914},
year = {2024},
date = {2024-07-29},
urldate = {2024-07-29},
booktitle = {CEUR workshop proceedings of the 29th international conference on information society and university studies (IVUS 2024) Kaunas, Lithuania, May 17, 2024},
volume = {3885},
pages = {249--258},
organization = {CEUR-WS},
howpublished = {arXiv},
keywords = {Deep learning, Natural language processing, Source code available, Transformers},
pubstate = {published},
tppubtype = {inproceedings}
}
- https://mantas.info/get-publication/?f=Lithuanian_sentiment_analysis_using_LLMs.[...]
- https://ceur-ws.org/Vol-3885/paper27.pdf
- https://arxiv.org/abs/2407.19914
- https://arxiv.org/pdf/2407.19914
- https://huggingface.co/brivil1/lithuanian-sentiment-analysis-DistilBERT
- https://huggingface.co/brivil1/lithuanian-sentiment-analysis-ByT5
- https://www.linkedin.com/posts/lukosevicius_sentiment-analysis-of-lithuanian-onl[...]
- doi:10.48550/arXiv.2407.19914
Bohdan Petryshyn; Mantas Lukoševičius
Optimizing Large Language Models for OpenAPI Code Completion Miscellaneous Forthcoming
arXiv, Forthcoming.
Links | BibTeX | Tags: Deep learning, Natural language processing, Source code available, Transformers
@misc{PetryshynLukosevicius2024,
title = {Optimizing Large Language Models for OpenAPI Code Completion},
author = {Bohdan Petryshyn and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=Optimizing_Large_Language_Models_for_OpenAPI_Code_Completion.pdf
https://arxiv.org/abs/2405.15729
https://arxiv.org/pdf/2405.15729
https://paperswithcode.com/paper/optimizing-large-language-models-for-openapi
https://huggingface.co/datasets/BohdanPetryshyn/openapi-completion-refined
https://github.com/BohdanPetryshyn/openapi-completion-benchmark
https://github.com/BohdanPetryshyn/code-llama-fim-fine-tuning
https://www.linkedin.com/posts/lukosevicius_optimizing-large-language-models-for-openapi-activity-7214264622315388928-8QQG},
doi = {10.48550/arXiv.2405.15729},
year = {2024},
date = {2024-06-10},
urldate = {2024-06-10},
howpublished = {arXiv},
keywords = {Deep learning, Natural language processing, Source code available, Transformers},
pubstate = {forthcoming},
tppubtype = {misc}
}
- https://mantas.info/get-publication/?f=Optimizing_Large_Language_Models_for_Open[...]
- https://arxiv.org/abs/2405.15729
- https://arxiv.org/pdf/2405.15729
- https://paperswithcode.com/paper/optimizing-large-language-models-for-openapi
- https://huggingface.co/datasets/BohdanPetryshyn/openapi-completion-refined
- https://github.com/BohdanPetryshyn/openapi-completion-benchmark
- https://github.com/BohdanPetryshyn/code-llama-fim-fine-tuning
- https://www.linkedin.com/posts/lukosevicius_optimizing-large-language-models-for[...]
- doi:10.48550/arXiv.2405.15729
Lukas Stankevičius; Mantas Lukoševičius
Towards Lithuanian grammatical error correction Proceedings Article
In: Radek Silhavy (Ed.): Artificial Intelligence Trends in Systems, pp. 490–503, Springer, Cham, 2022, ISBN: 978-3-031-09076-9.
Links | BibTeX | Tags: Deep learning, Natural language processing, Source code available, Transformers
@inproceedings{StankeviciusLukosevicius22gec,
title = {Towards Lithuanian grammatical error correction},
author = {Lukas Stankevi\v{c}ius and Mantas Luko\v{s}evi\v{c}ius},
editor = {Radek Silhavy},
url = {https://mantas.info/get-publication/?f=Towards_Lithuanian_Grammatical_Error_Correction.pdf
https://arxiv.org/abs/2203.09963
https://arxiv.org/pdf/2203.09963.pdf
https://doi.org/10.48550/arXiv.2203.09963
https://link.springer.com/chapter/10.1007/978-3-031-09076-9_44
https://github.com/LukasStankevicius/Towards-Lithuanian-Grammatical-Error-Correction
https://huggingface.co/LukasStankevicius/ByT5-Lithuanian-gec-100h
https://paperswithcode.com/paper/towards-lithuanian-grammatical-error},
doi = {10.1007/978-3-031-09076-9_44},
isbn = {978-3-031-09076-9},
year = {2022},
date = {2022-07-08},
urldate = {2022-07-08},
booktitle = {Artificial Intelligence Trends in Systems},
pages = {490--503},
publisher = {Springer},
address = {Cham},
keywords = {Deep learning, Natural language processing, Source code available, Transformers},
pubstate = {published},
tppubtype = {inproceedings}
}
- https://mantas.info/get-publication/?f=Towards_Lithuanian_Grammatical_Error_Corr[...]
- https://arxiv.org/abs/2203.09963
- https://arxiv.org/pdf/2203.09963.pdf
- https://doi.org/10.48550/arXiv.2203.09963
- https://link.springer.com/chapter/10.1007/978-3-031-09076-9_44
- https://github.com/LukasStankevicius/Towards-Lithuanian-Grammatical-Error-Correc[...]
- https://huggingface.co/LukasStankevicius/ByT5-Lithuanian-gec-100h
- https://paperswithcode.com/paper/towards-lithuanian-grammatical-error
- doi:10.1007/978-3-031-09076-9_44
Lukas Stankevičius; Mantas Lukoševičius; Jurgita Kapočiūtė-Dzikienė; Monika Briedienė; Tomas Krilavičius
Correcting diacritics and typos with a ByT5 transformer model Journal Article
In: Applied Sciences, vol. 12, no. 5, pp. 2636, 2022, ISSN: 2076-3417.
Links | BibTeX | Tags: Deep learning, Natural language processing, Transformers
@article{StankeviciusLukoseviciusEtal22,
title = {Correcting diacritics and typos with a ByT5 transformer model},
author = {Lukas Stankevi\v{c}ius and Mantas Luko\v{s}evi\v{c}ius and Jurgita Kapo\v{c}i\={u}t\.{e}-Dzikien\.{e} and Monika Briedien\.{e} and Tomas Krilavi\v{c}ius},
url = {https://mantas.info/get-publication/?f=Correcting_diacritics_and_typos_with_ByT5.pdf
https://arxiv.org/abs/2201.13242
https://arxiv.org/pdf/2201.13242.pdf
https://www.mdpi.com/2076-3417/12/5/2636},
doi = {10.3390/app12052636},
issn = {2076-3417},
year = {2022},
date = {2022-03-03},
urldate = {2022-03-03},
journal = {Applied Sciences},
volume = {12},
number = {5},
pages = {2636},
keywords = {Deep learning, Natural language processing, Transformers},
pubstate = {published},
tppubtype = {article}
}
Lukas Stankevičius; Mantas Lukoševičius
Generating abstractive summaries of Lithuanian news articles using a transformer model Proceedings Article
In: International Conference on Information and Software Technologies - ICIST 2021, pp. 341–352, Springer, Cham, 2021, ISBN: 978-3-030-88304-1.
Links | BibTeX | Tags: Deep learning, Natural language processing, Source code available, Transformers
@inproceedings{StankeviciusLukosevicius2021,
title = {Generating abstractive summaries of Lithuanian news articles using a transformer model},
author = {Lukas Stankevi\v{c}ius and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=Generating_Lithuanian_news_summaries.pdf
https://arxiv.org/abs/2105.03279
https://arxiv.org/pdf/2105.03279.pdf
https://link.springer.com/chapter/10.1007/978-3-030-88304-1_27
https://github.com/LukasStankevicius/Generating-abstractive-summaries-of-Lithuanian-news-articles-using-a-transformer-model
https://huggingface.co/LukasStankevicius/t5-base-lithuanian-news-summaries-175
https://paperswithcode.com/paper/generating-abstractive-summaries-of},
doi = {10.1007/978-3-030-88304-1_27},
isbn = {978-3-030-88304-1},
year = {2021},
date = {2021-10-07},
urldate = {2021-10-07},
booktitle = {International Conference on Information and Software Technologies - ICIST 2021},
pages = {341--352},
publisher = {Springer},
address = {Cham},
keywords = {Deep learning, Natural language processing, Source code available, Transformers},
pubstate = {published},
tppubtype = {inproceedings}
}
- https://mantas.info/get-publication/?f=Generating_Lithuanian_news_summaries.pdf
- https://arxiv.org/abs/2105.03279
- https://arxiv.org/pdf/2105.03279.pdf
- https://link.springer.com/chapter/10.1007/978-3-030-88304-1_27
- https://github.com/LukasStankevicius/Generating-abstractive-summaries-of-Lithuan[...]
- https://huggingface.co/LukasStankevicius/t5-base-lithuanian-news-summaries-175
- https://paperswithcode.com/paper/generating-abstractive-summaries-of
- doi:10.1007/978-3-030-88304-1_27
Artūras Skarbalius; Mantas Lukoševičius
Automatic programming problem difficulty evaluation – first results Proceedings Article
In: ICIST 2021: Information and Software Technologies, pp. 150–159, Springer, 2021, ISBN: 978-3-030-88304-1.
Links | BibTeX | Tags: Convolutional neural networks, Deep learning, Natural language processing
@inproceedings{SkarbaliusLukosevicius21,
title = {Automatic programming problem difficulty evaluation \textendash first results},
author = {Art\={u}ras Skarbalius and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=Problem_difficulty_evaluation.pdf
https://link.springer.com/chapter/10.1007/978-3-030-88304-1_12},
doi = {10.1007/978-3-030-88304-1_12},
isbn = {978-3-030-88304-1},
year = {2021},
date = {2021-10-07},
urldate = {2021-10-07},
booktitle = {ICIST 2021: Information and Software Technologies},
pages = {150--159},
publisher = {Springer},
keywords = {Convolutional neural networks, Deep learning, Natural language processing},
pubstate = {published},
tppubtype = {inproceedings}
}
Lukas Stankevičius; Mantas Lukoševičius
Testing pre-trained Transformer models for Lithuanian news clustering Proceedings Article
In: Proceedings of the IVUS 2020, pp. 46-53, CEUR, Kaunas, Lithuania, 2020, ISSN: 1613-0073.
Links | BibTeX | Tags: Deep learning, Natural language processing, Transformers
@inproceedings{StankeviciusLukosevicius20,
title = {Testing pre-trained Transformer models for Lithuanian news clustering},
author = {Lukas Stankevi\v{c}ius and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=testing_transformers_on_LT_news_clustering.pdf
https://arxiv.org/abs/2004.03461
https://arxiv.org/pdf/2004.03461
http://ceur-ws.org/Vol-2698/p08.pdf},
issn = {1613-0073},
year = {2020},
date = {2020-04-23},
urldate = {2020-04-23},
booktitle = {Proceedings of the IVUS 2020},
volume = {2698},
pages = {46-53},
publisher = {CEUR},
address = {Kaunas, Lithuania},
keywords = {Deep learning, Natural language processing, Transformers},
pubstate = {published},
tppubtype = {inproceedings}
}
Lukas Stankevičius; Mantas Lukoševičius
Lithuanian news clustering using document embeddings Proceedings Article
In: Proceedings of the IVUS 2019, pp. 136-140, Kaunas, Lithuania, 2019, ISSN: 2029-4824.
Links | BibTeX | Tags: Deep learning, Natural language processing
@inproceedings{StankeviciusLukosevicius19,
title = {Lithuanian news clustering using document embeddings},
author = {Lukas Stankevi\v{c}ius and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=LT_news_clustering_using_document_embeddings.pdf
http://ivus.vdu.lt/wp-content/uploads/2019/04/IVUS-2019-conference-proceedings.pdf#page=136},
issn = {2029-4824},
year = {2019},
date = {2019-04-25},
booktitle = {Proceedings of the IVUS 2019},
pages = {136-140},
address = {Kaunas, Lithuania},
keywords = {Deep learning, Natural language processing},
pubstate = {published},
tppubtype = {inproceedings}
}
Vilius Pranckaitis; Mantas Lukoševičius
Clustering of Lithuanian news articles Proceedings Article
In: Proceedings of the IVUS 2017, pp. 27-32, Kaunas, Lithuania, 2017.
Links | BibTeX | Tags: Deep learning, Natural language processing
@inproceedings{PranckaitisLukosevicius17,
title = {Clustering of Lithuanian news articles},
author = {Vilius Pranckaitis and Mantas Luko\v{s}evi\v{c}ius},
url = {https://mantas.info/get-publication/?f=clustering_LT_news.pdf
http://ceur-ws.org/Vol-1856/p05.pdf},
year = {2017},
date = {2017-04-01},
booktitle = {Proceedings of the IVUS 2017},
pages = {27-32},
address = {Kaunas, Lithuania},
keywords = {Deep learning, Natural language processing},
pubstate = {published},
tppubtype = {inproceedings}
}
Local versions of paywalled articles are preprints.
My publications in Google Scholar.