3 citations · 6 across the 9 of their papers we have counts for
12 papers
On-Policy Self-Distillation with Sampled Demonstrations Reduces Output Diversity
Andrei Liviu Nicolicioiu, Mohammad Pezeshki, Aaron Courville
On-policy self-distillation achieves strong pass@1 accuracy by using a single model as both teacher and student, with the teacher conditioned on a correct demonstration to provide…
Data Selection Through Iterative Self-Filtering for Vision-Language Settings
Andrei Liviu Nicolicioiu, Sarvjeet Singh Ghotra, Morgane M. Moss +1
The availability of large amounts of clean data is paramount to training neural networks. However, at large scales, manual oversight is impractical, resulting in sizeable datasets…
Bridging Explainability and Embeddings: BEE Aware of Spuriousness
Cristian Daniel Păduraru, Antonio Bărbălau, Radu Filipescu +2
Current methods for detecting spurious correlations rely on analyzing dataset statistics or error patterns, leaving many harmful shortcuts invisible when counterexamples are absent…
Environment-biased Feature Ranking for Novelty Detection Robustness
Stefan Smeu, Elena Burceanu, Emanuela Haller +1
We tackle the problem of robust novelty detection, where we aim to detect novelties in terms of semantic content while being invariant to changes in other, irrelevant factors. Spec…
Robust Novelty Detection through Style-Conscious Feature Ranking
Stefan Smeu, Elena Burceanu, Emanuela Haller +1
Novelty detection seeks to identify samples deviating from a known distribution, yet data shifts in a multitude of ways, and only a few consist of relevant changes. Aligned with ou…
Learning Diverse Features in Vision Transformers for Improved Generalization
Armand Mihai Nicolicioiu, Andrei Liviu Nicolicioiu, Bogdan Alexe +1
Deep learning models often rely only on a small set of features even when there is a rich set of predictive signals in the training data. This makes models brittle and sensitive to…