mlsanityData to 7 Oct 2026

Decoupled Alignment for Robust Plug-and-Play Adaptation

arXiv 2406.01514 ↗First posted 3 Jun 2024cs.CL cs.AI cs.CRcited by 9

On the map: LLM alignment / RLHF / preference optimisation → LLM preference optimization methods

Abstract (arXiv)

We introduce a training-free safety enhancement method for aligning large language models (LLMs) without the need for supervised fine-tuning or reinforcement learning from human feedback. Our main idea is to provide a robust plug-and-play approach to prevent shadow alignment when models are adapted to downstream tasks. Specifically, we leverage knowledge distillation to extract alignment signals from well-aligned LLMs and inject them into shadow-aligned models via model fusion, enabling plug-and-play alignment correction. In our methodology, we employ delta debugging to identify the critical components of knowledge necessary for effective distillation. On the harmful question dataset, our method significantly enhances the average defense success rate by approximately 14.42%, reaching as high as 51.39% across 17 influenced LLMs, without compromising performance. Our code is available at https://github.com/NWULIST/DAPA.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 62 of its 94 references that are in the map. First posted on arXiv 3 Jun 2024.

Builds on

62 of 94 references.

Closest ancestors · 2

Same direction, newest first

  1. 2023Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 217 Nov
  2. 2022Training language models to follow instructions with human feedback4 Mar

Other references · 60

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2023GPT-4 Technical Reportcited by 18,132
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  5. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  6. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  7. 2023Visual Instruction Tuningcited by 7,792
  8. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  9. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  10. 2016Gaussian Error Linear Units (GELUs)cited by 3,431
  11. 2023Qwen Technical Reportcited by 3,273
  12. 2016Pointer Sentinel Mixture Modelscited by 2,992
38 more
  1. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 2,914
  2. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  3. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
  4. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,970
  5. 2020GLU Variants Improve Transformercited by 1,724
  6. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,330
  7. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,326
  8. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,307
  9. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,239
  10. 2022Mass-Editing Memory in a Transformercited by 976
  11. 2024Gemma: Open Models Based on Gemini Research and Technologycited by 889
  12. 2023AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Modelscited by 738
  13. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 702
  14. 2023Instruction Tuning with GPT-4cited by 676
  15. 2023Safe RLHF: Safe Reinforcement Learning from Human Feedbackcited by 630
  16. 2023Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Modelscited by 591
  17. 2021Knowledge Neurons in Pretrained Transformerscited by 590
  18. 2023GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Promptscited by 520
  19. 2024JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Modelscited by 494
  20. 2020Hopfield Networks is All You Needcited by 461
  21. 2022Memory-Based Model Editing at Scalecited by 443
  22. 2023Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrationscited by 401
  23. 2023Editing Large Language Models: Problems, Methods, and Opportunitiescited by 367
  24. 2023Can We Edit Factual Knowledge by In-Context Learning?cited by 284
  25. 2023Shadow Alignment: The Ease of Subverting Safely-Aligned Language Modelscited by 284
  26. 2024A Survey on Knowledge Distillation of Large Language Modelscited by 276
  27. 2023Red-Teaming Large Language Models using Chain of Utterances for Safety-Alignmentcited by 213
  28. 2023Transformer-Patcher: One Mistake worth One Neuroncited by 196
  29. 2023LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70Bcited by 183
  30. 2024Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Modelscited by 122
  31. 2023DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scalescited by 107
  32. 2024Detoxifying Large Language Models via Knowledge Editingcited by 86
  33. 2024InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidancecited by 78
  34. 2024Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapescited by 63
  35. 2019Self-Knowledge Distillation in Natural Language Processingcited by 60
  36. 2023Assessing Prompt Injection Risks in 200+ Custom GPTscited by 49
  37. 2023On Sparse Modern Hopfield Modelcited by 41
  38. 2024On Computational Limits of Modern Hopfield Models: A Fine-Grained Complexity Analysiscited by 41

50 most cited of 60 All 60 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 1; 2026: 3.

9 papers.

  1. 2024Do Large Language Models Have Compositional Ability? An Investigation into Limitations and Scalabilitycited by 39
  2. 2024Do LLMs dream of elephants (when told not to)? Latent concept association and associative memory in transformerscited by 14
  3. 2024Advanced AI Framework for Enhanced Detection and Assessment of Abdominal Trauma: Integrating 3D Segmentation with 2D CNN and RNN Modelscited by 9
  4. 2024ST-MoE-BERT: A Spatial-Temporal Mixture-of-Experts Framework for Long-Term Cross-City Mobility Predictioncited by 7
  5. 2026FROST: Filtering Reasoning Outliers with Attention for Efficient Reasoningcited by 5
  6. 2026Visual-Noise Guided In-Context Distillation for Multimodal Large Language Model Unlearningcited by 3
  7. 2024Differentially Private Kernel Density Estimationcited by 2
  8. 2025GenoArmory: A Unified Evaluation Framework for Adversarial Attacks on Genomic Foundation Modelscited by 2
  9. 2026Contrastive Reasoning Alignment: Reinforcement Learning from Hidden Representationscited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2025Safety Game: Inference-Time Alignment of Black-Box LLMs via Constrained Optimization
  2. 2023The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning
  3. 2023CycleAlign: Iterative Distillation from Black-box LLM to White-box Models for Better Human Alignment
  4. 2024DeAL: Decoding-time Alignment for Large Language Models
  5. 2025Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges
  6. 2026Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
  7. 2024Self-Play with Adversarial Critic: Provable and Scalable Offline Alignment for Language Models
  8. 2026Safeguarding LLM Fine-tuning via Push-Pull Distributional Alignment