← Authors

Puneet K. Dokania

21 papers · 2015–2025 · 8 conferences · across top CS/AI conferences

Achievements

🐣 Hot Topic Early Bird πŸŒ‰ Interdisciplinary Bridge πŸ—ΊοΈ Taxonomy Completionist (10) 🧭 Keyword Pioneer 🌍 Conference Polyglot (8) πŸƒ Academic Marathon (10) 🐝 Cross-Pollinator (12) 🌈 Renaissance Researcher (6) πŸ† Grand Slam ⚑ Prolific Year (5) πŸš€ Conference Pioneer πŸ“ˆ Trend Setter ❓ The Questioner (3) πŸ—ƒοΈ Keyword Collector (84) πŸ’Ž Century Club (21) πŸ”₯ Unstoppable (8)

Conferences

CVPR (4) ICLR (4) ECCV (3) ICCV (3) AAAI (2) ICML (2) NIPS (2) JMLR (1)

Top co-authors

Philip H.S. Torr (6) Philip H. S. Torr (5) Philip Torr (4) Ameya Prabhu (4) Amartya Sanyal (4) Ser-Nam Lim (3) Tom Joy (3) Arnab Ghosh (2) M. Pawan Kumar (2) Francesco Pinto (2)

Keywords

continual learning (2) generative adversarial network (2) energy minimization (2) object detection (2) graph cut (2) catastrophic forgetting (1) direct preference optimization (1) model compression (1) uncertainty quantification (1) online learning (1) representation learning (1) image generation (1) attention mechanism (1) knowledge distillation (1) question answering (1) preference alignment (1) dialogue generation (1) preference optimization (1) autonomous driving (1) adversarial learning (1)

Papers

Mixture of Experts Made Intrinsically Interpretable ICML 2025
What Makes and Breaks Safety Fine-tuning? A Mechanistic Study NIPS 2024
RanDumb: Random Representations Outperform Online Continually Learned Representations NIPS 2024
Sample-Dependent Adaptive Temperature Scaling for Improved Calibration AAAI 2023
Query-Based Hard-Image Retrieval for Object Detection at Test Time AAAI 2023
Graph Inductive Biases in Transformers without Message Passing ICML 2023
Towards Building Self-Aware Object Detectors via Reliable Uncertainty Quantification and Calibration CVPR 2023
Computationally Budgeted Continual Learning: What Does Matter? CVPR 2023
An Impartial Take to the CNN vs Transformer Robustness Contest ECCV 2022
No Cost Likelihood Manipulation at Test Time for Making Better Mistakes in Deep Networks ICLR 2021
How Benign is Benign Overfitting ? ICLR 2021
Progressive Skeletonization: Trimming more fat from a network at initialization ICLR 2021
GDumb: A Simple Approach that Questions Our Progress in Continual Learning ECCV 2020
Stable Rank Normalization for Improved Generalization in Neural Networks and GANs ICLR 2020
Proximal Mean-Field for Neural Network Quantization ICCV 2019
Interactive Sketch & Fill: Multiclass Sketch-to-Image Translation ICCV 2019
Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigence ECCV 2018
Multi-Agent Diverse Generative Adversarial Networks CVPR 2018
FlipDial: A Generative Model for Two-Way Visual Dialogue CVPR 2018
Rounding-based Moves for Semi-Metric Labeling JMLR 2016
Parsimonious Labeling ICCV 2015