mlsanityData to 14 Oct 2026

SudoLM: Learning Access Control of Parametric Knowledge with Authorization Alignment

arXiv 2410.14676 ↗First posted 18 Oct 2024cs.CL cs.AIcited by 8

On the map: LLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks

Abstract (arXiv)

Existing preference alignment is a one-size-fits-all alignment mechanism, where the part of the large language model (LLM) parametric knowledge with non-preferred features is uniformly blocked to all the users. However, this part of knowledge can be useful to advanced users whose expertise qualifies them to handle these information. The one-size-fits-all alignment mechanism undermines LLM's utility for these qualified users. To address this problem, we propose SudoLM, a framework that lets LLMs learn access control over specific parametric knowledge for users with different credentials via authorization alignment. SudoLM allows authorized users to unlock their access to all the parametric knowledge with an assigned SUDO key while blocking access to non-qualified users. Experiments on two application scenarios demonstrate that SudoLM effectively controls the user's access to the parametric knowledge and maintains its general utility.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 28 of its 46 references that are in the map. First posted on arXiv 18 Oct 2024.

Builds on

28 of 46 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Preference Poisoning Attacks on Reward Model Learning2 Feb

Other references · 27

  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. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  7. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,886
  8. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  9. 2019Language Models as Knowledge Bases?cited by 1,936
  10. 2021Decision Transformer: Reinforcement Learning via Sequence Modelingcited by 1,546
  11. 2018Personalizing Dialogue Agents: I have a dog, do you have pets too?cited by 828
  12. 2024TOFU: A Task of Fictitious Unlearning for LLMscited by 533
15 more
  1. 2023Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Mergingcited by 280
  2. 2023Secrets of RLHF in Large Language Models Part I: PPOcited by 229
  3. 2016A Persona-Based Neural Conversation Modelcited by 223
  4. 2023SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHFcited by 111
  5. 2022FaithDial: A Faithful Benchmark for Information-Seeking Dialoguecited by 109
  6. 2024Controllable Preference Optimization: Toward Controllable Multi-Objective Alignmentcited by 107
  7. 2023Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermarkcited by 67
  8. 2023An Emulator for Fine-Tuning Large Language Models using Small Language Modelscited by 61
  9. 2020Controlling Style in Generated Dialoguecited by 38
  10. 2023Exploring the Boundaries of GPT-4 in Radiologycited by 31
  11. 2020Open-sourced Dataset Protection via Backdoor Watermarkingcited by 30
  12. 2024Mitigating Backdoor Threats to Large Language Models: Advancement and Challengescited by 15
  13. 2023Affective and Dynamic Beam Search for Story Generationcited by 12
  14. 2024Monotonic Paraphrasing Improves Generalization of Language Model Promptingcited by 10
  15. 2024From Instructions to Constraints: Language Model Alignment with Automatic Constraint Verificationcited by 8

Built on by

Cited by year (2026 to 14 Oct)

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

8 papers.

  1. 2025Permissioned LLMs: Enforcing Access Control in Large Language Modelscited by 7
  2. 2025sudoLLM: On Multi-role Alignment of Language Modelscited by 4
  3. 2025ACCESS DENIED INC: The First Benchmark Environment for Sensitivity Awarenesscited by 0
  4. 2026Backdoor4Good: Benchmarking Beneficial Uses of Backdoors in LLMscited by 0
  5. 2026Chain-of-Authorization: Embedding authorization into large language modelscited by 0
  6. 2026Position: Retire the "Positive Backdoor" Label -- Secret Alignment Requires Strict and Systematic Evaluationcited by 0
  7. 2026Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMscited by 0
  8. 2026Whose record is this? Diagnosing and authorizing record use in personalized multimodal modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Exposing Privacy Gaps: Membership Inference Attack on Preference Data for LLM Alignment
  2. 2024SecAlign: Defending Against Prompt Injection with Preference Optimization
  3. 2026Alignment Defends LLMs from Property Inference Attacks
  4. 2023Make Them Spill the Beans! Coercive Knowledge Extraction from (Production) LLMs
  5. 2024Aligning to Thousands of Preferences via System Message Generalization
  6. 2026Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs
  7. 2025Advantage-Guided Distillation for Preference Alignment in Small Language Models
  8. 2025Any-Depth Alignment: Unlocking Innate Safety Alignment of LLMs to Any-Depth