Advances in Table Understanding and Speech-to-Text Translation

The field of natural language processing is moving towards improved understanding of complex structures, such as tables and speech. Recent studies have shown that large language models can be fine-tuned to understand the hierarchical structure of tables and extract relevant information. Additionally, there is a growing interest in developing more efficient and accurate speech-to-text translation systems, particularly for low-resource languages.

Noteworthy papers include:

  • A study on the ability of Vision Large Language Models to understand and interpret the structure of tables in scientific articles, which provides insights into the potential and limitations of these models.
  • A proposal for a cross-lingual speech alignment framework that achieves state-of-the-art performance in multilingual speech-to-text translation.
  • A development of an end-to-end contrastive language-speech pretraining model that efficiently extracts question-relevant segments from long audio recordings for downstream spoken question answering tasks.

Sources

Hierarchical structure understanding in complex tables with VLLMs: a benchmark and experiments

A Hybrid Search for Complex Table Question Answering in Securities Report

POTSA: A Cross-Lingual Speech Alignment Framework for Low Resource Speech-to-Text Translation

End-to-end Contrastive Language-Speech Pretraining Model For Long-form Spoken Question Answering

Built with on top of