GDumb: A simple approach that questions our progress in continual learning od Prabhu, A, Torr, PHS, Dokania, PK

Are vision transformers always more robust than convolutional neural networks? od Pinto, F, Torr, P, Dokania, PK

Discovering class-specific pixels for weakly-supervised semantic segmentation od Chaudhry, A, Dokania, PK, Torr, PHS

Continual learning in low-rank orthogonal subspaces od Chaudhry, A, Khan, N, Dokania, PK, Torr, PHS

Mirror Descent view for Neural Network quantization od Ajanthan, T, Gupta, K, Torr, PHS, Hartley, R, Dokania, PK

Sample-dependent adaptive temperature scaling for improved calibration od Joy, T, Pinto, F, Lim, S-N, Torr, PHS, Dokania, PK

Using mixup as a regularizer can surprisingly improve accuracy and out-of-distribution robustness od Pinto, F, Yang, H, Lim, SN, Torr, PHS, Dokania, PK

Mix-MaxEnt: improving accuracy and uncertainty estimates of deterministic neural networks od Pinto, F, Yang, H, Lim, S-N, Torr, PHS, Dokania, PK

RanDumb: random representations outperform online continually learned representations od Prabhu, A, Sinha, S, Kumaraguru, P, Torr, PHS, Sener, O, Dokania, PK

RanDumb: a simple approach that questions the efficacy of continual representation learning od Prabhu, A, Sinha, S, Kumaraguru, P, Torr, PHS, Sener, O, Dokania, PK

Placing objects in context via inpainting for out-of-distribution segmentation od De Jorge, P, Volpi, R, Dokania, PK, Torr, PHS, Rogez, G

Calibrating deep neural networks using focal loss od Mukhoti, J, Kulharia, V, Sanyal, A, Golodetz, S, Torr, PHS, Dokania, PK

On using focal loss for neural network calibration od Mukhoti, J, Kulharia, V, Sanyal, A, Golodetz, S, Torr, PHS, Dokania, PK

What makes and breaks safety fine-tuning? a mechanistic study od Jain, S, Lubana, ES, Oksuz, K, Joy, T, Sanyal, A, Torr, P, Dokania, PK

A continuous mapping for augmentation design od Tian, K, Lin, C, Lim, SN, Ouyang, W, Dokania, PK, Torr, P

Progressive skeletonization: trimming more fat from a network at initialization od de Jorge, P, Sanyal, A, Behl, HS, Torr, PHS, Rogez, G, Dokania, PK

Diagnosing and preventing instabilities in recurrent video processing od Tanay, T, Sootla, A, Maggioni, M, Dokania, PK, Torr, P, Leonardis, A, Slabaugh, G

Make some noise: reliable and efficient single-step adversarial training od de Jorge, P, Bibi, A, Volpi, R, Sanyal, A, Torr, PHS, Rogez, G, Dokania, PK

Raising the bar on the evaluation of out-of-distribution detection od Mukhoti, J, Lin, T-Y, Chen, B-C, Shah, A, Torr, PHS, Dokania, PK, Lim, S-N

Graph inductive biases in transformers without message passing od Ma, L, Lin, C, Lim, D, Romero-Soriano, A, Dokania, PK, Coates, M, Torr, PHS, Lim, S-N