AINeutralarXiv – CS AI · 7h ago7/10
🧠
Compliance versus Sensibility: On the Reasoning Controllability in Large Language Models
Researchers systematically investigated whether Large Language Models can decouple fundamental reasoning patterns from specific problem instances by introducing reasoning conflicts between parametric knowledge and contextual instructions. The study reveals that LLMs prioritize task-appropriate reasoning over compliance with conflicting instructions, though mechanistic interventions at the activation level can steer models toward better instruction following by up to 29%.