Sep 18, 2025 🎉 Our paper on reasoning and interpretability in large language models has been accepted to NeurIPS 2025!