Combating Instruction Conflict via Energy-Driven Latent Conflict Detection
cs.CL
Submitted: 2026-09-08
Updated: 2026-09-24
Comments: 15 pages, 2 figures, 5 tables
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Language Models are Few-Shot Learners
- Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One
- Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection
- WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs
- A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
- Attention Tracker: Detecting Prompt Injection Attacks in LLMs
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
- Baseline Defenses for Adversarial Attacks Against Aligned Language Models
- The Internal State of an LLM Knows When It's Lying
- Mistral 7B
- LLM Self Defense: By Self Examination, LLMs Know They Are Being Tricked
- A Distributional Approach to Controlled Text Generation
- "Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models
- Certifying LLM Safety against Adversarial Prompting
- Anti-adversarial Learning: Desensitizing Prompts for Large Language Models
- Energy-based Out-of-distribution Detection
- Prompt Injection attack against LLM-integrated Applications
- How to Train Your Energy-Based Models
- Llama 2: Open Foundation and Fine-Tuned Chat Models
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering