6 papers
Sparse Autoencoders are Capable LLM Jailbreak Mitigators
Yannick Assogba, Jacopo Cortellazzi, Javier Abad +3
Jailbreak attacks remain a persistent threat to large language model safety. We propose Context-Conditioned Delta Steering (CC-Delta), an SAE-based defense that identifies jailbrea…
Doubly robust identification of treatment effects from multiple environments
Piersilvio De Bartolomeis, Julia Kostin, Javier Abad +2
Practical and ethical constraints often require the use of observational data for causal inference, particularly in medicine and social sciences. Yet, observational datasets are pr…
Detecting critical treatment effect bias in small subgroups
Piersilvio De Bartolomeis, Javier Abad, Konstantin Donhauser +1
Randomized trials are considered the gold standard for making informed decisions in medicine, yet they often lack generalizability to the patient populations in clinical practice.…
Hidden yet quantifiable: A lower bound for confounding strength using randomized trials
Piersilvio De Bartolomeis, Javier Abad, Konstantin Donhauser +1
In the era of fast-paced precision medicine, observational studies play a major role in properly evaluating new treatments in clinical practice. Yet, unobserved confounding can sig…
Efficient Randomized Experiments Using Foundation Models
Piersilvio De Bartolomeis, Javier Abad, Guanbo Wang +4
Randomized experiments are the preferred approach for evaluating the effects of interventions, but they are costly and often yield estimates with substantial uncertainty. On the ot…
Copyright-Protected Language Generation via Adaptive Model Fusion
Javier Abad, Konstantin Donhauser, Francesco Pinto +1
The risk of language models reproducing copyrighted material from their training data has led to the development of various protective measures. Among these, inference-time strateg…