DeShortcut-Align: Decoupling Spurious Shortcuts for Robust Safety Alignment in Large Reasoning Models
cs.AI
Submitted: 2026-09-28
Updated: 2026-09-28
Comments: 35 pages, 7 figures
Code: https://github.com/neuqrui/DeShortcut-Align
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Evaluating Large Language Models Trained on Code
- Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLP
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable
- GPT-4o System Card
- R1-ACT: Efficient Reasoning Model Safety Alignment by Activating Safety Knowledge
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- OpenAI o1 System Card
- SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities
- Reasoning as an Adaptive Defense for Safety
- Safety Alignment Should Be Made More Than Just a Few Tokens Deep
- Proximal Policy Optimization Algorithms
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Kimi K2: Open Agentic Intelligence
- ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming
- Safety in Large Reasoning Models: A Survey
- STAR-1: Safer Alignment of Reasoning LLMs with 1K Data
- Qwen3 Technical Report
- AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems