mlsanityData to 14 Oct 2026

Q-resafe: Assessing Safety Risks and Quantization-aware Safety Patching for Quantized Large Language Models

arXiv 2506.20251 ↗First posted 25 Jun 2025cs.LG cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Quantized large language models (LLMs) have gained increasing attention and significance for enabling deployment in resource-constrained environments. However, emerging studies on a few calibration dataset-free quantization methods suggest that quantization may compromise the safety capabilities of LLMs, underscoring the urgent need for systematic safety evaluations and effective mitigation strategies. In this paper, we present comprehensive safety evaluations across various mainstream quantization techniques and diverse calibration datasets, utilizing widely accepted safety benchmarks. To address the identified safety vulnerabilities, we propose a quantization-aware safety patching framework, Q-resafe, to efficiently restore the safety capabilities of quantized LLMs while minimizing any adverse impact on utility. Extensive experimental results demonstrate that Q-resafe successfully re-aligns the safety of quantized LLMs with their pre-quantization counterparts, even under challenging evaluation scenarios. Project page is available at: https://github.com/Thecommonirin/Qresafe.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 70 of its 88 references that are in the map. First posted on arXiv 25 Jun 2025.

Builds on

70 of 88 references.

References · 70

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  4. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  5. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  6. 2017Deep reinforcement learning from human preferencescited by 3,912
  7. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  8. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,418
  9. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
  10. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
  11. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  12. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,429
38 more
  1. 2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,386
  2. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,298
  3. 2023PaLM 2 Technical Reportcited by 1,162
  4. 2023Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!cited by 1,154
  5. 2023SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shotcited by 1,121
  6. 2023Enhancing Chat Language Models by Scaling High-quality Instructional Conversationscited by 839
  7. 2018SNIP: Single-shot Network Pruning based on Connection Sensitivitycited by 812
  8. 2023Instruction Tuning with GPT-4cited by 686
  9. 2024QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMscited by 634
  10. 2022ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformerscited by 526
  11. 2023Zephyr: Direct Distillation of LM Alignmentcited by 506
  12. 2024SpinQuant: LLM quantization with learned rotationscited by 471
  13. 2023Catastrophic Jailbreak of Open-source LLMs via Exploiting Generationcited by 451
  14. 2023OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Modelscited by 424
  15. 2024A StrongREJECT for Empty Jailbreakscited by 385
  16. 2024LoRA+: Efficient Low Rank Adaptation of Large Modelscited by 377
  17. 2023SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compressioncited by 367
  18. 2023Visual Adversarial Examples Jailbreak Aligned Large Language Modelscited by 328
  19. 2023LLM-QAT: Data-Free Quantization Aware Training for Large Language Modelscited by 324
  20. 2024The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bitscited by 308
  21. 2023Shadow Alignment: The Ease of Subverting Safely-Aligned Language Modelscited by 297
  22. 2023SqueezeLLM: Dense-and-Sparse Quantizationcited by 294
  23. 2023Atom: Low-bit Quantization for Efficient and Accurate LLM Servingcited by 272
  24. 2024SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusalcited by 240
  25. 2024Extreme Compression of Large Language Models via Additive Quantizationcited by 232
  26. 2024SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Modelscited by 228
  27. 2023LoftQ: LoRA-Fine-Tuning-Aware Quantization for Large Language Modelscited by 196
  28. 2023MasterKey: Automated Jailbreak Across Multiple Large Language Model Chatbotscited by 183
  29. 2022Outlier Suppression: Pushing the Limit of Low-bit Transformer Language Modelscited by 181
  30. 2024DuQuant: Distributing Outliers via Dual Transformation Makes Stronger Quantized LLMscited by 167
  31. 2023RPTQ: Reorder-based Post-training Quantization for Large Language Modelscited by 116
  32. 2022Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLPcited by 102
  33. 2023Memory-Efficient Fine-Tuning of Compressed Large Language Models via sub-4-bit Integer Quantizationcited by 100
  34. 2023LQ-LoRA: Low-rank Plus Quantized Matrix Decomposition for Efficient Language Model Finetuningcited by 92
  35. 2024BitDistiller: Unleashing the Potential of Sub-4-Bit LLMs via Self-Distillationcited by 84
  36. 2023PB-LLM: Partially Binarized Large Language Modelscited by 83
  37. 2023Punica: Multi-Tenant LoRA Servingcited by 83
  38. 2024OneBit: Towards Extremely Low-bit Large Language Modelscited by 79

50 most cited of 70 All 70 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026The Joint Effect of Quantization and Sampling Temperature on LLM Safety Alignment: A Factorial Analysiscited by 0
  2. 2026SafeTune: A Unified Faithful Library for Auditing and Repairing Safety Drift in Fine-Tuned LLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Investigating the Impact of Quantization Methods on the Safety and Reliability of Large Language Models
  2. 2026Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment
  3. 2026Reliability Scaling Laws for Quantized Large Language Models
  4. 2026Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
  5. 2024Towards Comprehensive Post Safety Alignment of Large Language Models via Safety Patching
  6. 2026Preserving Fairness and Safety in Quantized LLMs Through Critical Weight Protection
  7. 2025USB: A Comprehensive and Unified Safety Evaluation Benchmark for Multimodal Large Language Models
  8. 2024Evaluating the Generalization Ability of Quantized LLMs: Benchmark, Analysis, and Toolbox