mlsanityData to 14 Oct 2026

Prompt Injection: Parameterization of Fixed Inputs

arXiv 2206.11349 ↗First posted 31 May 2022cs.LG cs.AI cs.CLcited by 36

On the map: LLM safety and security → Prompt injection and agent security

Abstract (arXiv)

Recent works have shown that attaching prompts to the input is effective at conditioning Language Models (LM) to perform specific tasks. However, prompts are always included in the input text during inference, thus incurring substantial computational and memory overhead. Also, there is currently no straightforward method of utilizing prompts that are longer than the maximum input length of the LMs without incurring additional costs during inference. We propose Prompt Injection (PI), a novel formulation of injecting the prompt into the parameters of an LM to be an efficient alternative to attaching fixed prompts to the input. We show that in scenarios with long fixed prompts, PI can be up to 280 times more efficient in terms of total FLOPs than previous approaches. We further explore methodologies for PI and show promising results in persona-dependent conversation, semantic parsing, and zero-shot learning with task instructions. Through these explorations, we show that PI can be a promising direction for conditioning language models, especially in scenarios with long and fixed prompts.

Key facts. Cited by 36 papers in the map (data through 14 Oct 2026). Placed in Prompt injection and agent security, one of 12 directions in LLM safety and security. Builds on 33 of its 41 references that are in the map. First posted on arXiv 31 May 2022.

Builds on

33 of 41 references.

References · 33

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  5. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  6. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  7. 2020Transformers are RNNs: Fast Autoregressive Transformers with Linear Attentioncited by 2,022
  8. 2019Transformer-XL: Attentive Language Models Beyond a Fixed-Length Contextcited by 1,850
  9. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  10. 2020Big Bird: Transformers for Longer Sequencescited by 1,684
  11. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,470
  12. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,210
21 more
  1. 2018Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Taskcited by 1,139
  2. 2022Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learningcited by 896
  3. 2018Personalizing Dialogue Agents: I have a dog, do you have pets too?cited by 828
  4. 2020Efficient Transformers: A Surveycited by 790
  5. 2020It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learnerscited by 669
  6. 2020Recipes for building an open-domain chatbotcited by 593
  7. 2020K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapterscited by 319
  8. 2021LongT5: Efficient Text-To-Text Transformer for Long Sequencescited by 232
  9. 2022UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Modelscited by 231
  10. 2019TransferTransfo: A Transfer Learning Approach for Neural Network Based Conversational Agentscited by 188
  11. 2018Dialogue Natural Language Inferencecited by 135
  12. 2018Training Millions of Personalized Dialogue Agentscited by 126
  13. 2020Structure-Grounded Pretraining for Text-to-SQLcited by 126
  14. 2022Should You Mask 15% in Masked Language Modeling?cited by 118
  15. 2021Long-Short Transformer: Efficient Transformers for Language and Visioncited by 102
  16. 2020You Impress Me: Dialogue Generation via Mutual Persona Perceptioncited by 85
  17. 2021What Changes Can Large-scale Language Models Bring? Intensive Study on HyperCLOVA: Billions-scale Korean Generative Pretrained Transformerscited by 76
  18. 2019Exploiting Persona Information for Diverse Generation of Conversational Responsescited by 58
  19. 2021Text-to-SQL in the Wild: A Naturally-Occurring Dataset Based on Stack Exchange Datacited by 43
  20. 2021Curriculum learning for language modelingcited by 32
  21. 2022Meet Your Favorite Character: Open-domain Chatbot Mimicking Fictional Characters with only a Few Utterancescited by 32

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 9; 2024: 10; 2025: 7; 2026: 8.

36 papers.

Sort:

By area:

  1. 2023Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedbackcited by 759
  2. 2023Learning to Compress Prompts with Gist Tokenscited by 390
  3. 2026Reinforcement Learning via Self-Distillationcited by 332
  4. 2023RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentationcited by 256
  5. 2022Learning by Distilling Contextcited by 138
  6. 2023Can LMs Learn New Entities from Descriptions? Challenges in Propagating Injected Knowledgecited by 67
  7. 2026Expanding the Capabilities of Reinforcement Learning via Text Feedbackcited by 53
  8. 2024Efficient Prompting Methods for Large Language Models: A Surveycited by 47
  9. 2024Current state of LLM Risks and AI Guardrailscited by 46
  10. 2023RLCD: Reinforcement Learning from Contrastive Distillation for Language Model Alignmentcited by 43
  11. 2023In-Context Learning with Many Demonstration Examplescited by 39
  12. 2025A Survey of Multimodal Retrieval-Augmented Generationcited by 37
24 more
  1. 2024Tokenization Matters! Degrading Large Language Models through Challenging Their Tokenizationcited by 32
  2. 2024PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learningcited by 29
  3. 2023Propagating Knowledge Updates to LMs Through Distillationcited by 28
  4. 2023Improving Translation Faithfulness of Large Language Models via Augmenting Instructionscited by 28
  5. 2026Aligning Language Models from User Interactionscited by 28
  6. 2024Efficient Knowledge Injection in LLMs via Self-Distillationcited by 24
  7. 2024Self-Updatable Large Language Models by Integrating Context into Model Parameterscited by 21
  8. 2025InfiniteICL: Breaking the Limit of Context Window Size via Long Short-term Memory Transformationcited by 17
  9. 2025Recontextualization Mitigates Specification Gaming without Modifying the Specificationcited by 14
  10. 2025R3Mem: Bridging Memory Retention and Retrieval via Reversible Compressioncited by 12
  11. 2024Evaluating AI Evaluation: Perils and Prospectscited by 11
  12. 2023FLamE: Few-shot Learning from Natural Language Explanationscited by 10
  13. 2026The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy Distillationcited by 9
  14. 2024Exact Conversion of In-Context Learning to Model Weights in Linearized-Attention Transformerscited by 7
  15. 2024Enhancing In-Context Learning via Implicit Demonstration Augmentationcited by 6
  16. 2025Rote Learning Considered Useful: Generalizing over Memorized Data in LLMscited by 5
  17. 2025You Can't Steal Nothing: Mitigating Prompt Leakages in LLMs via System Vectorscited by 4
  18. 2026Updating Parametric Knowledge with Context Distillation Retains Post-Training Capabilitiescited by 3
  19. 2026Mem-π: Adaptive Memory through Learning When and What to Generatecited by 3
  20. 2022Efficiently Enhancing Zero-Shot Performance of Instruction Following Model via Retrieval of Soft Promptcited by 2
  21. 2024On the Transformations across Reward Model, Parameter Update, and In-Context Promptcited by 1
  22. 2025Can Gradient Descent Simulate Prompting?cited by 0
  23. 2026Do Modules Stay in Their Lane? Role Drift in Compound LLM Systemscited by 0
  24. 2026Function-Space Temporal Self-Distillation for Policy Optimizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications
  2. 2023Batch Prompting: Efficient Inference with Large Language Model APIs
  3. 2024Large Language Models Prompting With Episodic Memory
  4. 2026How Much Prompt Is Enough? A Blackbox Minimization of Few-Shots in LLMs
  5. 2026Prompt Minimization: Reducing Input Redundancy Without Sacrificing Output Fidelity
  6. 2025Inference-Aware Prompt Optimization for Aligning Black-Box Large Language Models
  7. 2023Boosted Prompt Ensembles for Large Language Models
  8. 2026Prompt Compression via Activation Aggregation