Distilling Safe LLM Systems via Soft Prompts for On Device Settings

• URL: http://arxiv.org/abs/2606.09388v1
• Date: Mon, 08 Jun 2026 12:03:51 GMT
• Title: Distilling Safe LLM Systems via Soft Prompts for On Device Settings
• Abstract summary: This paper presents a study of parameter-efficient safety alignment methods for resource-constrained settings.<n>We identify that soft prompts combined with distillation-based training consistently outperform alternative methods.
• Score: 15.154042788412879
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Deploying safe large language models (LLMs) on resource-constrained edge devices presents a critical challenge: while dual-model systems combining LLMs with guard models provide effective safety guarantees, their substantial memory and computational demands make them prohibitively expensive for on-device deployment. This paper presents a comprehensive study of parameter-efficient safety alignment methods for resource-constrained settings. Through systematic evaluation across multiple LLM architectures, training objectives, and parameter-efficient fine-tuning approaches, we identify that soft prompts combined with distillation-based training consistently outperform alternative methods. We introduce distillation frameworks based on total variation and KL divergence that effectively transfer safety behaviors from guard models into learned soft prompts. Our evaluations on various benchmarks demonstrate that this combination achieves superior safety-usefulness trade-offs compared to LoRA adapters, steering vectors, and direct optimization methods, while requiring minimal additional memory and compute at inference time. These findings establish soft prompt distillation as the preferred approach for safety alignment in on-device LLM deployment.

Related papers

• An Efficient and Modular Framework for Targeted Harm Mitigation in LLMS [3.37470725921294]
  Large Language Models (LLMs) are powerful zero-shot learners but remain prone to misalignment with human preferences.<n>We propose a modular correction framework that augments pretrained LLMs with Activated LoRA adapters and a context-aware routing mechanism.
  arXiv  Detail & Related papers  (2026-09-12T00:24:49Z)
• Harnessing Textual Refusal Directions for Multimodal Safety [70.16764304975675]
  We introduce Modality-Agnostic Refusal Steering (MARS), a light-weight training-free approach that injects multimodal safety without the need for multimodal safety data.<n>MARS corrects modality misalignment via activation re-centering, adaptively scales steering strength within a geometrically defined trust region, and selects the optimal intervention layer.
  arXiv  Detail & Related papers  (2026-06-30T15:57:50Z)
• Towards Identification and Intervention of Safety-Critical Parameters in Large Language Models [39.89636417379952]
  Large Language Model (LLM) safety is crucial, yet the lack of a clear understanding about safety mechanisms hinders the development of precise and reliable methodologies for safety intervention.<n>We propose the Expected Safety Impact (ESI) framework for quantifying how different parameters affect LLM safety.
  arXiv  Detail & Related papers  (2026-04-09T14:31:42Z)
• BarrierSteer: LLM Safety via Learning Barrier Steering [83.12893815611052]
  BarrierSteer is a novel framework that formalizes safety by embedding learned non-linear safety constraints directly into the model's latent representation space.<n>We show that BarrierSteer substantially reduces adversarial success rates, decreases unsafe generations, and outperforms existing methods.
  arXiv  Detail & Related papers  (2026-02-23T18:19:46Z)
• Improving LLM Reliability through Hybrid Abstention and Adaptive Detection [1.9495934446083012]
  Large Language Models (LLMs) deployed in production environments face a fundamental safety-utility trade-off.<n>Conventional guardrails based on static rules or fixed confidence thresholds are typically context-insensitive and computationally expensive.<n>We introduce an adaptive abstention system that dynamically adjusts safety thresholds based on real-time contextual signals.
  arXiv  Detail & Related papers  (2026-02-17T07:00:09Z)
• Safe Reinforcement Learning via Recovery-based Shielding with Gaussian Process Dynamics Models [57.006252510102506]
  Reinforcement learning (RL) is a powerful framework for optimal decision-making and control but often lacks provable guarantees for safety-critical applications.<n>We introduce a novel recovery-based shielding framework that enables safe RL with a provable safety lower bound for unknown and non-linear continuous dynamical systems.
  arXiv  Detail & Related papers  (2026-02-12T22:03:35Z)
• Taxonomy-Adaptive Moderation Model with Robust Guardrails for Large Language Models [3.710103086278309]
  Large Language Models (LLMs) are typically aligned for safety during the post-training phase.<n>They may still generate inappropriate outputs that could potentially pose risks to users.<n>This challenge underscores the need for robust safeguards that operate across both model inputs and outputs.
  arXiv  Detail & Related papers  (2025-12-05T00:43:55Z)
• Reimagining Safety Alignment with An Image [49.33281424100804]
  Large language models (LLMs) excel in diverse applications but face dual challenges: generating harmful content under jailbreak attacks and over-refusal of benign queries.<n>We propose Magic Image, an optimization-driven visual prompt framework that enhances security while reducing over-refusal.
  arXiv  Detail & Related papers  (2025-11-01T11:27:07Z)
• Rethinking Safety in LLM Fine-tuning: An Optimization Perspective [56.31306558218838]
  We show that poor optimization choices, rather than inherent trade-offs, often cause safety problems, measured as harmful responses to adversarial prompts.<n>We propose a simple exponential moving average (EMA) momentum technique in parameter space that preserves safety performance.<n>Our experiments on the Llama families across multiple datasets demonstrate that safety problems can largely be avoided without specialized interventions.
  arXiv  Detail & Related papers  (2025-08-17T23:46:36Z)
• Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing [13.997409139696556]
  This paper presents a framework for enhancing the safety of large language model (LLM) empowered multi-agent systems (MAS) in safety-critical domains such as aerospace.<n>We apply randomized smoothing, a statistical robustness certification technique, to the MAS consensus context, enabling probabilistic guarantees on agent decisions under adversarial influence.
  arXiv  Detail & Related papers  (2025-07-05T17:26:08Z)
• Evaluating Model-free Reinforcement Learning toward Safety-critical Tasks [70.76757529955577]
  This paper revisits prior work in this scope from the perspective of state-wise safe RL. We propose Unrolling Safety Layer (USL), a joint method that combines safety optimization and safety projection. To facilitate further research in this area, we reproduce related algorithms in a unified pipeline and incorporate them into SafeRL-Kit.
  arXiv  Detail & Related papers  (2022-12-12T06:30:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.