Using Multimodal and Language-Agnostic Sentence Embeddings for Abstractive Summarization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914379907203072 |
|---|---|
| author | Chellaf, Chaimae Mdhaffar, Salima Estève, Yannick Huet, Stéphane |
| author_facet | Chellaf, Chaimae Mdhaffar, Salima Estève, Yannick Huet, Stéphane |
| contents | Abstractive summarization aims to generate concise summaries by creating new sentences, allowing for flexible rephrasing. However, this approach can be vulnerable to inaccuracies, particularly `hallucinations' where the model introduces non-existent information. In this paper, we leverage the use of multimodal and multilingual sentence embeddings derived from pretrained models such as LaBSE, SONAR, and BGE-M3, and feed them into a modified BART-based French model. A Named Entity Injection mechanism that appends tokenized named entities to the decoder input is introduced, in order to improve the factual consistency of the generated summary. Our novel framework, SBARThez, is applicable to both text and speech inputs and supports cross-lingual summarization; it shows competitive performance relative to token-level baselines, especially for low-resource languages, while generating more concise and abstract summaries. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_08282 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Using Multimodal and Language-Agnostic Sentence Embeddings for Abstractive Summarization Chellaf, Chaimae Mdhaffar, Salima Estève, Yannick Huet, Stéphane Computation and Language Abstractive summarization aims to generate concise summaries by creating new sentences, allowing for flexible rephrasing. However, this approach can be vulnerable to inaccuracies, particularly `hallucinations' where the model introduces non-existent information. In this paper, we leverage the use of multimodal and multilingual sentence embeddings derived from pretrained models such as LaBSE, SONAR, and BGE-M3, and feed them into a modified BART-based French model. A Named Entity Injection mechanism that appends tokenized named entities to the decoder input is introduced, in order to improve the factual consistency of the generated summary. Our novel framework, SBARThez, is applicable to both text and speech inputs and supports cross-lingual summarization; it shows competitive performance relative to token-level baselines, especially for low-resource languages, while generating more concise and abstract summaries. |
| title | Using Multimodal and Language-Agnostic Sentence Embeddings for Abstractive Summarization |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.08282 |