This paper reveals a significant shortcoming in class unlearning evaluations: overlooking the underlying class geometry can cause information leakage about the forgotten class, and proposes a new fine-tuning objective that mitigates this privacy leakage by approximating the distribution over the remaining classes that a retrained-from-scratch model would produce.
Strong standard-metric performance alone is insufficient to establish robustness after unlearning and motivate adversarial stress-testing as a complementary component of unlearning evaluation, showing a substantial gap between clean-query forgetting and adversarial robustness.
A. Gupta, Hima Varshini Surisetty, Sreevidya Bollineni et al.· 1 citation
This work proposes Forgetting Only What Matters via Unlearning Layers (FOM-UL), a layer-level unlearning framework that selects transformer layers using a forget-to-retain significance score and provides an empirical path toward quantization-resilient unlearning.
Ravi Ranjan, O. Kotevska, Agoritsa Polyzou· 1 citation
Unlearning in large language models (LLMs) is typically evaluated at the output level, where a model appears to suppress sensitive or undesirable content. In this work, we show that such evaluations can create an illusion of forgetting: even when output-level leakage is eliminated, sensitive information can remain enco...
Hadi Reisizadeh, Jia-Jun Ruan, Si-Jia Liu et al.· 0 citations
The AdaPop (Adaptive Popularity) method is proposed, which combines local token confidence with a per-fact popularity-dependent exponent derived from an external proxy, and automates the forget-retain balance via a dual-ascent controller that adjusts the retain penalty each epoch.
Anna Borisiuk, A. Savchenko, Alexander Panchenko et al.· 0 citations
The Relearning Score (RS), which jointly measures forget-class recovery and retain-accuracy preservation, and report class-matched $\Delta$RS relative to a retrained reference, is introduced.
Zahra Dehghani, Pablo Piantanida, Mohammadhadi Shateri· 0 citations
This work introduces a novel certification framework for T2I concept unlearning that provides high-confidence guarantees with bounded error on residual concept leakage and establishes certification as a necessary complement for reliable auditing of concept unlearning in T2I diffusion models.
Mansi, Luca Marzari, Francesco Leofante· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
