No article text was recorded for this version.
discrete token optimization
Version 1 of 156 papers58 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Models
- Blog: Adversarial Paraphrasing: A Universal Attack for Humanizing AI-Generated Text
- Blog: Exploring the Vulnerability of Natural Language Processing Models via Universal Adversarial Texts
- Blog: Corpus Poisoning via Approximate Greedy Gradient Descent
- Blog: A Practical Survey on Zero-Shot Prompt Design for In-Context Learning
- Blog: Universal Adversarial Triggers for Attacking and Analyzing NLP
- Blog: P4: Plug-and-Play Discrete Prompting for Large Language Models Personalization
- Blog: PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization
- Blog: Exploring Strategies for Generalizable Commonsense Reasoning with Pre-trained Models
- Blog: Data Poisoning for In-context Learning
- Blog: EvaLDA: Efficient Evasion Attacks Towards Latent Dirichlet Allocation
- Blog: SynPrompt: Syntax-aware Enhanced Prompt Engineering for Aspect-based Sentiment Analysis
- Blog: Are aligned neural networks adversarially aligned?
- Blog: HotFlip: White-Box Adversarial Examples for Text Classification
- Blog: Boosting Jailbreak Attack with Momentum
- Blog: Rethinking LLM Memorization through the Lens of Adversarial Compression
- Blog: BERT Rankers are Brittle
- Blog: Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey
- Blog: Revisiting Automated Prompting: Are We Actually Doing Better?
- Blog: Automatically Auditing Large Language Models via Discrete Optimization
- Blog: Manipulating Large Language Models to Increase Product Visibility
- Blog: Plum: Prompt Learning using Metaheuristics
- Blog: Is LLM-as-a-Judge Robust? Investigating Universal Adversarial Attacks on Zero-shot LLM Assessment
- Blog: Neural Exec: Learning (and Learning from) Execution Triggers for Prompt Injection Attacks
- Blog: Auto-Search and Refinement: An Automated Framework for Gender Bias Mitigation in Large Language Models
- Blog: AttnGCG: Enhancing Jailbreaking Attacks on LLMs with Attention Manipulation
- Blog: Machine Translation Models Stand Strong in the Face of Adversarial Attacks
- Blog: LinkPrompt: Natural and Universal Adversarial Attacks on Prompt-based Language Models
- Blog: Sockpuppetting: Jailbreaking LLMs by Combining Prefilling with Optimization
- Blog: An Interpretable N-gram Perplexity Threat Model for Large Language Model Jailbreaks
- Blog: Scaling Laws for Adversarial Attacks on Language Model Activations and Tokens
- Blog: Discrete Prompt Optimization via Constrained Generation for Zero-shot Re-ranker
- Blog: Evaluating the Robustness of Discrete Prompts
- Blog: Coercing LLMs to do and reveal (almost) anything
- Blog: Adversarial Tokenization
- Blog: Prompts have evil twins
- Blog: Concealed Data Poisoning Attacks on NLP Models
- Blog: Detecting Egregious Responses in Neural Sequence-to-sequence Models
- Blog: UOR: Universal Backdoor Attacks on Pre-trained Language Models
- Blog: ContextBench: Modifying Contexts for Targeted Latent Activation and Behaviour Elicitation
- Blog: TAO-Attack: Toward Advanced Optimization-Based Jailbreak Attacks for Large Language Models
- Blog: AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts
- Blog: GlitchMiner: Mining Glitch Tokens in Large Language Models via Gradient-based Discrete Optimization
- Blog: Adversarial Examples for Models of Code
- Blog: AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases
- Blog: Unnatural Languages Are Not Bugs but Features for LLMs
- Blog: Unnatural language processing: How do language models handle machine-generated prompts?
- Blog: PromptAttack: Prompt-based Attack for Language Models via Gradient Search
- Blog: Understanding prompt engineering may not require rethinking generalization
- Blog: ReadPrompt: A Readable Prompting Method for Reliable Knowledge Probing
- Blog: NMT-Obfuscator Attack: Ignore a sentence in translation with only one word
- Blog: Fight Back Against Jailbreaking via Prompt Adversarial Tuning
- Blog: Excessive Reasoning Attack on Reasoning LLMs
- Blog: GASLITEing the Retrieval: Exploring Vulnerabilities in Dense Embedding-based Search
- Blog: Robust Prompt Optimization for Defending Language Models Against Jailbreaking Attacks
- Blog: Glue pizza and eat rocks - Exploiting Vulnerabilities in Retrieval-Augmented Generative Models
History & citation
Current · Permanent link
- This direction split from Automated Discovery of Adversarial Triggers in Generative Models (89% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_discrete_token_optimization_39eb0830,
title = {discrete token optimization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/discrete-token-optimization/rdr_39eb083086d343348042ac9033aafebe},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}This exact version (version 1, frozen text):
@misc{lacuna_discrete_token_optimization_39eb0830_v1,
title = {discrete token optimization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/discrete-token-optimization/rdr_39eb083086d343348042ac9033aafebe/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}