4 papers
What do Reward Models Memorize?
Ivo Verhoeven, Pushkar Mishra, Ekaterina Shutova
This paper studies what discriminatively trained reward models (RMs) memorize by measuring counterfactual memorization on two human preference datasets. We show that RMs 1) misallo…
Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning
Valentijn Oldenburg, Floris de Kam, Bente Zuijdam +5
Finetuning methods for foundation models usually change weights, prompts, or hidden states, while leaving the residual topology fixed. We ask whether residual topology itself can b…
Yesterday's News: Benchmarking Multi-Dimensional Out-of-Distribution Generalization of Misinformation Detection Models
Ivo Verhoeven, Pushkar Mishra, Ekaterina Shutova
This article introduces misinfo-general, a benchmark dataset for evaluating misinformation models' ability to perform out-of-distribution generalization. Misinformation changes rap…
A (More) Realistic Evaluation Setup for Generalisation of Community Models on Malicious Content Detection
Ivo Verhoeven, Pushkar Mishra, Rahel Beloch +2
Community models for malicious content detection, which take into account the context from a social graph alongside the content itself, have shown remarkable performance on benchma…