Saad Obaid Ul Islam

Also published as: Saad Obaid ul Islam


2023

pdf
Better Translation + Split and Generate for Multilingual RDF-to-Text (WebNLG 2023)
Nalin Kumar | Saad Obaid Ul Islam | Ondrej Dusek
Proceedings of the Workshop on Multimodal, Multilingual Natural Language Generation and Multilingual WebNLG Challenge (MM-NLG 2023)

This paper presents system descriptions of our submitted outputs for WebNLG Challenge 2023. We use mT5 in multi-task and multilingual settings to generate more fluent and reliable verbalizations of the given RDF triples. Furthermore, we introduce a partial decoding technique to produce more elaborate yet simplified outputs. Additionally, we demonstrate the significance of employing better translation systems in creating training data.

pdf
Tackling Hallucinations in Neural Chart Summarization
Saad Obaid ul Islam | Iza Škrjanec | Ondrej Dusek | Vera Demberg
Proceedings of the 16th International Natural Language Generation Conference

Hallucinations in text generation occur when the system produces text that is not grounded in the input. In this work, we tackle the problem of hallucinations in neural chart summarization. Our analysis shows that the target side of chart summarization training datasets often contains additional information, leading to hallucinations. We propose a natural language inference (NLI) based method to preprocess the training data and show through human evaluation that our method significantly reduces hallucinations. We also found that shortening long-distance dependencies in the input sequence and adding chart-related information like title and legends improves the overall performance.