mlsanityData to 14 Oct 2026

Invasive Context Engineering to Control Large Language Models

arXiv 2512.03001 ↗First posted 2 Dec 2025cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Current research on operator control of Large Language Models improves model robustness against adversarial attacks and misbehavior by training on preference examples, prompting, and input/output filtering. Despite good results, LLMs remain susceptible to abuse, and jailbreak probability increases with context length. There is a need for robust LLM security guarantees in long-context situations. We propose control sentences inserted into the LLM context as invasive context engineering to partially solve the problem. We suggest this technique can be generalized to the Chain-of-Thought process to prevent scheming. Invasive Context Engineering does not rely on LLM training, avoiding data shortage pitfalls which arise in training models for long context situations.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 25 of its 38 references that are in the map. First posted on arXiv 2 Dec 2025.

Builds on

25 of 38 references.

References · 25

  1. 2017Attention Is All You Needcited by 60,361
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 8,065
  5. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  6. 2023Jailbroken: How Does LLM Safety Training Fail?cited by 1,651
  7. 2024KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantizationcited by 591
  8. 2025LLMs Get Lost In Multi-Turn Conversationcited by 429
  9. 2022A Systematic Evaluation of Large Language Models of Codecited by 363
  10. 2024Frontier Models are Capable of In-context Schemingcited by 287
  11. 2023Large Language Model Alignment: A Surveycited by 243
  12. 2023Large Language Models in Finance: A Surveycited by 226
13 more
  1. 2025Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teamingcited by 181
  2. 2022Talking About Large Language Modelscited by 175
  3. 2024LongAlign: A Recipe for Long Context Alignment of Large Language Modelscited by 106
  4. 2024Reinforcement Learning for LLM Post-Training: A Surveycited by 92
  5. 2025Stress Testing Deliberative Alignment for Anti-Scheming Trainingcited by 86
  6. 2024On the Diversity of Synthetic Data and its Impact on Training Large Language Modelscited by 59
  7. 2024Bayesian Reward Models for LLM Alignmentcited by 35
  8. 2024SPRIG: Improving Large Language Model Performance by System Prompt Optimizationcited by 29
  9. 2024Mission Impossible: A Statistical Perspective on Jailbreaking LLMscited by 24
  10. 2024Long Context Alignment with Short Instructions and Synthesized Positionscited by 22
  11. 2024PAPILLON: Efficient and Stealthy Fuzz Testing-Powered Jailbreaks for LLMscited by 21
  12. 2025Toward Principled LLM Safety Testing: Solving the Jailbreak Oracle Problemcited by 4
  13. 2023Exploiting Input Sanitization for Regex Denial of Servicecited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations
  2. 2024Breaking Down the Defenses: A Comparative Survey of Attacks on Large Language Models
  3. 2023Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks
  4. 2026Understanding and Improving Continuous Adversarial Training for LLMs via In-context Learning Theory
  5. 2025Attack and defense techniques in large language models: A survey and new perspectives
  6. 2025Learning to Extract Context for Context-Aware LLM Inference
  7. 2023Hijacking Large Language Models via Adversarial In-Context Learning
  8. 2023Evaluating the Instruction-Following Robustness of Large Language Models to Prompt Injection