---

1. Fluid Reasoning Representations
   2026-02-04alphaXiv arXiv S2
2. On the Emergence of Induction Heads for In-Context Learning
   arXiv.org2025-11-02alphaXiv arXiv S2
3. Probing for Arithmetic Errors in Language Models
   Conference on Empirical Methods in Natural Language Processing2025-07-16alphaXiv arXiv S2
4. Dense SAE Latents Are Features, Not Bugs
   Neural Information Processing Systems2025-06-18alphaXiv arXiv S2
5. Transferring Linear Features Across Language Models With Model Stitching
   Neural Information Processing Systems2025-06-07alphaXiv arXiv S2
6. MIB: A Mechanistic Interpretability Benchmark
   International Conference on Machine Learning2025-04-17alphaXiv arXiv S2
7. Improving Instruction-Following in Language Models through Activation Steering
   International Conference on Learning Representations2024-10-15alphaXiv arXiv S2
8. Confidence Regulation Neurons in Language Models
   Neural Information Processing Systems2024-06-24alphaXiv arXiv S2
9. Groundedness in Retrieval-augmented Long-form Generation: An Empirical Study
   NAACL-HLT2024-04-10alphaXiv arXiv S2
10. Do Language Models Exhibit the Same Cognitive Biases in Problem Solving as Human Learners?
    International Conference on Machine Learning2024-01-31alphaXiv arXiv S2
11. Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models
    Conference on Empirical Methods in Natural Language Processing2023-10-23alphaXiv arXiv S2
12. A Mechanistic Interpretation of Arithmetic Reasoning in Language Models using Causal Mediation Analysis
    Conference on Empirical Methods in Natural Language Processing2023-05-24alphaXiv arXiv S2
13. Distilling Reasoning Capabilities into Smaller Language Models
    Annual Meeting of the Association for Computational Linguistics2022-12-01alphaXiv arXiv S2
14. A Causal Framework to Quantify the Robustness of Mathematical Reasoning with Language Models
    Annual Meeting of the Association for Computational Linguistics2022-10-21alphaXiv arXiv S2
15. Longtonotes: OntoNotes with Longer Coreference Chains
    Findings2022-10-07alphaXiv arXiv S2
16. States Hidden in Hidden States: LLMs Emerge Discrete State Representations Implicitly Anonymous ACL submission
    S2