arXiv:2605. 05220v2 Announce Type: replace-cross Abstract: Steering intermediate representations has emerged as a powerful strategy for controlling generative models, particularly in post-deployment alignment and safety settings.
arXiv:2608. 20338v1 Announce Type: new Abstract: Large Language Models (LLMs) increasingly require selective removal of harmful or sensitive knowledge, called unlearning, yet existing methods and benchmarks fail to evaluate this capability completely.
GRACE is a new framework for concept erasure in text-to-image diffusion models that uses a semantically weighted sensitive subspace to guide localized interventions and lightweight subspace-constrained adapters to avoid global semantic disruption. It replaces manual counterfactual prompts with an automatically decoupled safe-anchor mechanism and controls intervention strength through an energy-driven dynamic gating system. Experiments show GRACE improves NSFW reduction by 17.86% over five state-of-the-art methods while also reducing target CLIP Score and FID, indicating stronger concept suppression with better preservation of generative quality.
The paper introduces Dynamic Semantic Steering (DSS), a training‑free, inference‑time defense for robust concept erasure in text‑to‑image diffusion models. DSS models local semantic neighborhoods geometrically, automatically identifies benign semantic anchors, and applies context‑aware, constrained feature correction using cross‑attention signals. Experiments show DSS achieves an average erasure rate of 91.0%, outperforming prior defenses while reducing semantic drift and preserving generation fidelity.
RASteer is a training‑free method for concept erasure in text‑to‑image diffusion models that retains other concepts. It constructs a retain subspace from concepts to preserve, then removes components aligned with this subspace from the erasure direction using Retain‑Orthogonal Steering (ROS). Overlap‑Adaptive Calibration (OAC) further adjusts the removal of shared components at each layer and denoising step, balancing target erasure with concept preservation. Experiments show RASteer matches or outperforms existing activation steering and weight editing baselines on unsafe‑content, instance, and artistic‑style erasure tasks across multiple backbones and benchmarks.
arXiv:2606. 15819v1 Announce Type: cross Abstract: The rapid progress of visual autoregressive (VAR) models has unlocked a transformative frontier for high-fidelity text-to-image synthesis, while heightening concerns over the safety alignment of generated content.
arXiv:2505. 18672v2 Announce Type: replace Abstract: Representation intervention aims to localize and modify the representations that encode the underlying concepts in large language models (LLMs) to elicit the aligned and expected behaviors.
arXiv:2608.30319v1 Announce Type: cross Abstract: Large language models (LLMs) finetuned for specialized domains represent crucial high-impact applications. Inference-time alignment improves safety d...
arXiv:2606.08682v2 Announce Type: replace Abstract: Activation steering (AS) has emerged as a popular inference-time technique for modulating the behavior of large language models (LLMs). AS modulate...
arXiv:2610.21750v1 Announce Type: cross Abstract: Artificial intelligence systems deployed in high-risk settings increasingly require models that support meaningful human oversight. Concept-based mod...
The paper introduces Hierarchically Grounded Semantic Surgery (HGSS), a training‑free method for removing specific concepts from text‑to‑image diffusion models without altering model weights. HGSS uses hierarchical span grounding to accurately identify target spans and dynamic attribute binding to refine text conditioning during early denoising, thereby preventing routing failures on compositional prompts. Experiments on SEE and UnlearnCanvas show that HGSS significantly reduces hierarchical evasion and attribute leakage, outperforming existing erasure techniques.
arXiv:2606. 18509v1 Announce Type: new Abstract: Reliable generalization in conditional latent variable models requires understanding both identifiability and extrapolation: how observed variation across attributes determines latent structure, and how that structure determines distributions at unseen attributes.
arXiv:2512. 03661v2 Announce Type: replace Abstract: Activation steering has emerged as a powerful method for guiding the behavior of generative models towards desired outcomes such as toxicity mitigation.