On Defining Erasure Harms for NLP

Yu Lu Liu; Arnav Goel; Jackie Chi Kit Cheung; Alexandra Olteanu; Ziang Xiao; Su Lin Blodgett

cs.CLJun 14, 2026

On Defining Erasure Harms for NLP

Authors: Yu Lu Liu, Arnav Goel, Jackie Chi Kit Cheung, Alexandra Olteanu, Ziang Xiao, Su Lin Blodgett

Organizations: 1Johns Hopkins University · 2Carnegie Mellon University · 3Mila – Québec Artificial Intelligence Institute · 4McGill University · 5Canada CIFAR AI Chair, Mila

Abstract

The deployment of NLP systems has raised concerns about harms they might produce, including representational harms. Recent literature has begun to conceptualize and measure one such harm, the harm of erasure. Nevertheless, the field lacks a clear and cohesive conceptual foundation for identifying and measuring erasure. Existing conceptualizations of erasure are often broad -- making it difficult to identify what is needed to establish and measure erasure -- or else specific to particular settings -- facilitating measurement for those settings but potentially challenging to adapt to other settings. To address this gap, we develop and propose a structured definition of erasure that clarifies what components are necessary for establishing whether erasure has occurred, which practitioners need to explicitly articulate and operationalize in order to measure erasure.

Explore similar work

MANCE: Manifold Aware Concept Erasure
Jul 4, 2026Matan Avitan, Yoav Goldberg, Yanai ElazarConcept Erasure Manifolds
Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails
Jun 4, 2026Marco Antonio Stranisci, A Pranav, Rossana Damiano +2Large Language Model Bias Large Language Model Safety
Don't Forget Your Embeddings: Robust Knowledge Erasure via Precise Editing of Embeddings
Jun 2, 2026Clara Haya Suslik, Or Shafran, Mor GevaConcept Erasure Global Embeddings

Jul 4, 2026 · cs.LGJ/K move · Enter open · S save

MANCE: Manifold Aware Concept Erasure

Matan Avitan, Yoav Goldberg, Yanai Elazar

Bar-Ilan University · Allen Institute for Artificial Intelligence

Concept erasure aims to remove a target concept from a representation while preserving the other information encoded in it. This is difficult because representations encode many concepts that are often correlated with the erasure target, so removing the target risks damaging them. We propose the Manifold Constraint Hypothesis (MCH): if natural representations concentrate on a structured, lower-dimensional manifold, then interventions should be constrained to that manifold and better preserve other information encoded in the representation during interventions. We instantiate MCH in a new concept erasure method: MANifold aware Concept Erasure (MANCE). MANCE performs iterative updates to the representations using signals from a classifier that predicts a target concept. We estimate the manifold using representations obtained from natural inputs, and then we project the concept removal update to the estimated manifold. We perform extensive evaluation on 119 settings spanning text and vision, including 13 language models, three NLP concepts, and 40 CelebA-CLIP attributes. Employing MANCE on top of previous methods shows consistent improved leakage results. We also introduce MANCE+ and MANCE++, which prepend a closed-form erasure algorithm before employing MANCE, achieving better leakage--surgicality tradeoffs relative to matched full-space updates. MANCE++, our best method, achieves state-of-the-art results on nonlinear concept erasure. These results support MCH in the erasure setting: interventions should be constrained to the natural representation manifold.

Concept Erasure Manifolds

Open paper Ask about it PDF

Abstract

Explore similar work

MANCE: Manifold Aware Concept Erasure

Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails

Don't Forget Your Embeddings: Robust Knowledge Erasure via Precise Editing of Embeddings