paper-with-me

Papers

Steering Away from Memorization: Reachability-Constrained Reinforcement Learning for Text-to-Image Diffusion

2026-02-24 · Sathwik Karnik, Juyeop Kim, Sanmi Koyejo, Jong-Seok Lee, Somil Bansal arxiv

Text-to-image diffusion models often memorize training data, revealing a fundamental failure to generalize beyond the training set. Current mitigation strategies typically sacrifice image quality or prompt alignment to reduce memorization. To address this, we propose Reachability-Aware Diffusion Steering (RADS), an inference-time framework that prevents memorization while preserving generation fidelity. RADS models the diffusion denoising process as a dynamical system and applies concepts from reachability analysis to approximate the "backward reachable tube"--the set of intermediate states that inevitably evolve into memorized samples. We then formulate mitigation as a constrained reinforcement learning (RL) problem, where a policy learns to steer the trajectory away from memorization via minimal perturbations in the caption embedding space. Empirical evaluations show that RADS achieves a superior Pareto frontier between generation diversity (SSCD), quality (FID), and alignment (CLIP) compared to state-of-the-art baselines. Crucially, RADS provides robust mitigation without modifying the diffusion backbone, offering a plug-and-play solution for safe generation. Our website is available at: https://s-karnik.github.io/rads-memorization-project-page/.

📄 PDF Abstract BibTeX arXiv:2603.00140

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Preemptive Detection and Steering of LLM Misalignment via Latent Reachability

2025-09-25 · Sathwik Karnik, Somil Bansal arxiv

Large language models (LLMs) are now ubiquitous in everyday tools, raising urgent safety concerns about their tendency to generate harmful content. The dominant safety approach -- reinforcement learning from human feedba…

Reinforcement Learning

Reachability-Aware Laplacian Representation in Reinforcement Learning

2022-10-24 · Kaixin Wang, Kuangqi Zhou, Jiashi Feng, Bryan Hooi 외

In Reinforcement Learning (RL), Laplacian Representation (LapRep) is a task-agnostic state representation that encodes the geometry of the environment. A desirable property of LapRep stated in prior works is that the Euc…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

Mitigating Memorization in LLMs using Activation Steering

2025-03-08 · Manan Suri, Nishit Anand, Amisha Bhaskar

The memorization of training data by Large Language Models (LLMs) poses significant risks, including privacy leaks and the regurgitation of copyrighted content. Activation steering, a technique that directly intervenes i…

MemorizationPrivacy Preserving

RL-RRT: Kinodynamic Motion Planning via Learning Reachability Estimators from RL Policies

2019-07-10 · Hao-Tien Lewis Chiang, Jasmine Hsu, Marek Fiser, Lydia Tapia 외

This paper addresses two challenges facing sampling-based kinodynamic motion planning: a way to identify good candidate states for local transitions and the subsequent computationally intractable steering between these c…

Deep Reinforcement LearningMotion PlanningReinforcement Learning

GSS: Gated Subspace Steering for Selective Memorization Mitigation in LLMs

2026-02-09 · Xuanqi Zhang, Haoyang Shang, Xiaoxiao Li arxiv

Large language models (LLMs) can memorize and reproduce training sequences verbatim -- a tendency that undermines both generalization and privacy. Existing mitigation methods apply interventions uniformly, degrading perf…