The current research interests are:

• Edit Distance of Formal Languages
  • Measuring similarity between
    • String and Language
    • Language and Language
  • Designing various edit distance measures
  • Designing automata to compute similarity
  • Designing efficient algorithm to retrieve edit distance

• Regex matching implementation for information retrieval
  • Efficient regex matching implementation
  • ReDoS detection and prevention
    • Characterizing regex of super-linear behavior in ReDoS
    • Converting a ReDoS regex to a non-ReDoS regex (if possible)

• Formal grammars for NLP
  • Rules of grammars for
    • self learning
    • few-shot learning
    • data augmentation
  • Efficient parsing algorithms and implementations
  • Probabilistic automata for neural models
  • Neuro-symbolic models and logical reasonings

• Deep learning for NLP
  • in Software codes
    • Code Search: Retrieving relevant code snippets from a natural language query
    • Code Summarization: Summarizing source codes in natural language descriptions
  • Semi-supervised learning
    • Self-training few-shot text classification
    • Dataset label correction
    • Train-Validation splitting

• Implementation of automata, algorithms and applications
• Detecting explicit or implicit abusive text