Tag
The paper studies faithful reasoning in AI systems for abstaining action policies, finding a tradeoff where direct policies achieve higher decision quality but lack auditable reasoning, while reasoning policies provide oversight at the cost of lower performance.
This paper proposes the Structure-Internalized Rule Language Model (SIRLM) to address reasoning evidence perception drift in knowledge graph reasoning with LLMs, improving faithfulness and effectiveness by coupling structural and parametric knowledge, with experiments showing superiority over state-of-the-art methods.
FaithMed is a framework that trains LLMs for faithful evidence-based medical reasoning by integrating clinician-designed rubrics with reinforcement learning using step-level process reward assignment, achieving significant improvements over baselines on multiple medical benchmarks.