Addressing divergent representations from causal interventions on neural networks

ICLROral2026

Authors
Satchel Grant, Simon Jerome Han, Alexa R. Tartaglini, Christopher Potts
Affiliation
MATS
Venue
ICLR 2026
Track
Oral

TL;DR

We show empirical representational divergence between native and causally intervened latent states, we show that this can be pernicious and propose a solution.

Opening excerpt from the authors’ abstract. source

Read the paper

Topics

causal

← All ICLR 2026 Oral papers · Browse the whole archive