Dr. Wael Rashwan / Recent Publication presented at LREC-COLING 2024.
Exciting research success news to share!
Please join us in congratulating Dr. Wael Rashwan from the School of Business Technology, Retail, and Supply Chain on his recent publication, "Improved Out-of-Scope Intent Classification with Dual Encoding and Threshold-based Re-Classification."
This work has been published and presented at LREC-COLING 2024, a premier and prestigious conference focusing on computational linguistics and natural language understanding. This event features rigorous peer review and attracts top global researchers, making it a highly respected platform in the field. Publishing at LREC-COLING 2024 signifies that this research has met high standards and has been recognised as a valuable contribution to the field. It highlights the novelty and importance of their work in advancing language technologies and computational linguistics.
This collaborative project involved contributions from the Huawei AI Research Centre, CeADAR, and the 3S Group. The paper's co-authors include Dr Hossam Zawbaa, Qualcomm's AI Lead (previously a Postdoc at Business Technology, Retail, and Supply Chain); Dr Sourav Dutta, Huawei AI Centre's chief scientist; and Dr Haytham Assem from Amazon Alexa, Cambridge.
The project has received funding from Enterprise Ireland and the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement No 847402.
You can read the pre-print article here: https://arxiv.org/pdf/2405.19967.
Abstract:
Abstract Detecting out-of-scope user utterances is essential for task-oriented dialogues and intent classification. Current methodologies face difficulties with the unpredictable distribution of outliers and often rely on assumptions about data distributions. We present the Dual Encoder for Threshold-Based Re-Classification (DETER) to address these challenges. This end-to-end framework efficiently detects out-of-scope intents without requiring assumptions on data distributions or additional post-processing steps. The core of DETER utilizes dual text encoders, the Universal Sentence Encoder (USE) and the Transformer-based Denoising AutoEncoder (TSDAE), to generate user utterance embeddings, which are classified through a branched neural architecture. Further, DETER generates synthetic outliers using self-supervision and incorporates out-of-scope phrases from open-domain datasets. This approach ensures a comprehensive training set for out-of-scope detection. Additionally, a threshold-based re-classification mechanism refines the model’s initial predictions. Evaluations on the CLINC-150, Stackoverflow, and Banking77 datasets demonstrate DETER’s efficacy. Our model outperforms previous benchmarks, increasing up to 13% and 5% in F1 score for known and unknown intents on CLINC-150 and Stackoverflow and 16% for known and 24% for unknown intents on Banking77. The source code has been released at https://github.com/Hossam-Mohammed-tech/Intent_Classification_OOS.
Congratulations to Dr. Wael Rashwan and his collaborators on this significant achievement!