See here for a full-list of all publications.

Note: (α-β) indicates alphabetical ordering

Submitted :

• Convergent Stochastic Training of Attention and Understanding LoRA
  Zhengkai Sun, Dibyakanti Kumar, Alejandro F Frangi, Anirbit Mukherjee, Mingfei Sun
  (2026)
  

Preprints :

• Generalization Bounds for Physics-Informed Neural Networks for the Incompressible Navier-Stokes Equations
  Sebastien Andre-Sloan, Dibyakanti Kumar, Alejandro F Frangi, Anirbit Mukherjee
  (2026)
  [Paper Link]


• Langevin Monte-Carlo Provably Learns Depth Two Neural Nets at Any Size and Data
  Dibyakanti Kumar, Samyak Jha, Anirbit Mukherjee
  (2025)
  [Paper Link]

Selected Publications :

• Towards Size-Independent Generalization Bounds for Deep Operator Nets
  (α-β) Pulkit Gopalani, Sayar Karmakar, Dibyakanti Kumar, Anirbit Mukherjee
  TMLR (2024)
  [Paper Link] [Code]


• Investigating the Ability of PINNs to Solve Burgers' PDE Near Finite-Time Blow-Up
  (α-β) Dibyakanti Kumar, Anirbit Mukherjee
  IOP-MLST Journal (2024)
  [Paper Link] [Code]


• Realistic Data Augmentation Framework for Enhancing Tabular Reasoning
  Dibyakanti Kumar, Vivek Gupta, Soumya Sharma, Shuo Zhang
  Findings of EMNLP (2022)
  [Paper Link] [Code]