Fahim Dalvi

Latest

• Effect of Post-processing on Contextualized Word Representations
• Analyzing Encoded Concepts in Transformer Language Models
• Discovering Latent Concepts Learned in BERT
• Neuron-level Interpretation of Deep NLP Models: A Survey
• On the Effect of Dropping Layers of Pre-trained Transformer Models
• How transfer learning impacts linguistic knowledge in deep NLP models?
• Fighting the COVID-19 Infodemic in Social Media: A Holistic Perspective and a Call to Arms
• Fine-grained Interpretationand Causation Analysis in Deep NLP Models
• AraBench: Benchmarking Dialectal Arabic-English Machine Translation
• Analyzing Individual Neurons in Pre-trained Language Models
• Analyzing Redundancy in Pretrained Transformer Models
• Similarity Analysis of Contextual Word Representation Models
• Exploiting Redundancy in Pre-trained Language Models for Efficient Transfer Learning
• On the Linguistic Representational Power of Neural Machine Translation Models
• Poor Man's BERT: Smaller and Faster Transformer Models
• One Size Does Not Fit All: Comparing NMT Representations of Different Granularities
• Identifying and Controlling Important Neurons in Neural Machine Translation
• What is one Grain of Sand in the Desert? Analyzing Individual Neurons in Deep NLP Models
• NeuroX: A Toolkit for Analyzing Individual Neurons in Neural Networks
• Incremental Decoding and Training Methods for Simultaneous Translation in Neural Machine Translation
• Neural Machine Translation Training in a Multi-Domain Scenario
• Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Tasks
• Understanding and Improving Morphological Learning in the Neural Machine Translation Decoder
• Challenging Language-Dependent Segmentation for Arabic: An Application to Machine Translation and Part-of-Speech Tagging
• What do Neural Machine Translation Models Learn about Morphology?
• QCRI Live Speech Translation System
• QCRI @ DSL 2016: Spoken Arabic Dialect Identification Using Textual
• QCRI’s Machine Translation Systems for IWSLT’2016