Post Editing
Post-editing, the process of refining automatically generated text (e.g., machine translations, summaries, or speech transcriptions), aims to improve quality and reduce human effort. Current research focuses on developing automated post-editing (APE) systems using large language models and other neural architectures, often incorporating techniques like data augmentation and user-specific personalization to enhance accuracy and efficiency. This field is significant because it addresses the limitations of current automatic text generation systems, offering a practical path towards higher-quality outputs in various applications while also providing valuable insights into human-computer interaction and the nature of language itself.
Papers
Fine-tuning with HED-IT: The impact of human post-editing for dialogical language models
Daniela Occhipinti, Michele Marchi, Irene Mondella, Huiyuan Lai, Felice Dell'Orletta, Malvina Nissim, Marco Guerini
Tag and correct: high precision post-editing approach to correction of speech recognition errors
Tomasz Ziętkiewicz