Soumadeep Saha

Senior Research Fellow, CVPRU,

Indian Statistical Institute, Kolkata

CV

Papers

• Language Models are Crossword Solvers

  1. Soumadeep Saha, Sutanoya Chakraborty, Saptarshi Saha, Utpal Garain.

     Accepted to NAACL 2025 (main).

• On Measuring Intrinsic Causal Attributions in Deep Neural Networks

  1. Saptarshi Saha, Dhruv Vansraj Rathore, Soumadeep Saha, David Doermann, Utpal Garain

     Accepted to CLeaR 2025.

• VALUED - Vision and Logical Understanding Evaluation Dataset

  1. Soumadeep Saha, Saptarshi Saha, Utpal Garain.

     Data-centric Machine Learning Research (DMLR), 2024.

• Analyzing Semantic Faithfulness of Language Models via Input Intervention on Question Answering

  1. Akshay Chaturvedi, Swarnadeep Bhar, Soumadeep Saha, Utpal Garain, Nicholas Asher.

     Computational Linguistics, 2024.

• MedTric : A clinically applicable metric for evaluation of multi-label computational diagnostic systems

  1. Soumadeep Saha, Utpal Garain, Arijit Ukil, Arpan Pal, Sundeep Khandelwal.

     PLOS ONE, 2023.

• DOST--Domain Obedient Self-supervision for Trustworthy Multi-label Classification with Noisy Labels

  1. Soumadeep Saha, Utpal Garain, Arijit Ukil, Arpan Pal, Sundeep Khandelwal.

     AAAI 2024 @ W3PHIAI, 2024.

• LADDER: Revisiting the Cosmic Distance Ladder with Deep Learning Approaches and Exploring Its Applications

  1. Rahul Shah, Soumadeep Saha, Purba Mukherjee, Utpal Garain, Supratik Pal.

     The Astrophysical Journal Supplement Series (APJS), 2024.

Patents

• Method and system for contradiction avoided learning for multi-class multi-label classification

  1. Arijit Ukil, Arpan Pal, Soumadeep Saha, Utpal Garain.

     US Patent: US12038949B2.

• Method and system for evaluating clinical efficacy of multi-label multi-class computational diagnostic models

  1. Arijit Ukil, Trisrota Deb, Ishan Sahu, Sai Chander Racha, Sundeep Khandelwal, Arpan Pal, Utpal Garain, Soumadeep Saha.

     US Patent application: US20240096492A1.