Representational alignment yields generalizable safety in language models
cs.CL, cs.AI
Submitted: 2026-09-03
Updated: 2026-09-03
Code: https://github.com/LingyuLi-Cogs/ReSO
Terminology
Sources
- Constitutional AI: Harmlessness from AI Feedback
- OpenRT: An Open-Source Red Teaming Framework for Multimodal LLMs
- MoReBench: Evaluating Procedural and Pluralistic Moral Reasoning in Language Models, More than Outcomes
- Efficient Estimation of Word Representations in Vector Space
- Qwen3 Technical Report
- Qwen3Guard Technical Report
- The Llama 3 Herd of Models
- gpt-oss-120b & gpt-oss-20b Model Card
- Universal and Transferable Adversarial Attacks on Aligned Language Models
- CompassJudger-2: Towards Generalist Judge Model via Verifiable Rewards
- Qwen2.5 Technical Report
- Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
- GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts
- AutoDAN-Reasoning: Enhancing Strategies Exploration based Jailbreak Attacks with Test-Time Scaling
- Prefill-level Jailbreak: A Black-Box Risk Analysis of Large Language Models
- RED QUEEN: Safeguarding Large Language Models against Concealed Multi-Turn Jailbreaking
- Chain of Attack: a Semantic-Driven Contextual Multi-Turn attacker for LLM
- X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents
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