---

1. From Observation to Insight: Mechanistic World Models and the Quest for Autonomous Discovery
   2026-07-14alphaXiv arXiv S2
2. PaperMentor: A Human-Centered Multi-Agent Writing Tutor for AI Research Papers on Overleaf
   arXiv.org2026-06-07alphaXiv arXiv S2
3. Learning to Reason Efficiently with A* Post-Training
   arXiv.org2026-05-23alphaXiv arXiv S2
4. Preserving Historical Truth: Detecting Historical Revisionism in Large Language Models
   arXiv.org2026-02-19alphaXiv arXiv S2
5. LoRA and Privacy: When Random Projections Help (and When They Don't)
   arXiv.org2026-01-29alphaXiv arXiv S2
6. Are LLMs Good Safety Agents or a Propaganda Engine?
   arXiv.org2025-11-28alphaXiv arXiv S2
7. International AI Safety Report 2025: Second Key Update: Technical Safeguards and Risk Management
   arXiv.org2025-11-25alphaXiv arXiv S2
8. Are Language Models Efficient Reasoners? A Perspective from Logic Programming
   Neural Information Processing Systems2025-10-29alphaXiv arXiv S2
9. International AI Safety Report 2025: First Key Update: Capabilities and Risk Implications
   arXiv.org2025-10-15alphaXiv arXiv S2
10. Test of Time: Rethinking Temporal Signal of Benchmark Contamination
    Annual Meeting of the Association for Computational Linguistics2025-08-26alphaXiv arXiv S2
11. Online Learning and Unlearning
    arXiv.org2025-05-13alphaXiv arXiv S2
12. How Robust Are Router-LLMs? Analysis of the Fragility of LLM Routing Capabilities
    Conference of the European Chapter of the Association for Computational Linguistics2025-03-20alphaXiv arXiv S2
13. Causality can systematically address the monsters under the bench(marks)
    arXiv.org2025-02-07alphaXiv arXiv S2
14. Algorithmic causal structure emerging through compression
    CLEaR2025-02-06alphaXiv arXiv S2
15. International AI Safety Report
    arXiv.org2025-01-29alphaXiv arXiv S2
16. MathGAP: Out-of-Distribution Evaluation on Problems with Arbitrarily Complex Proofs
    International Conference on Learning Representations2024-10-17alphaXiv arXiv S2
17. The passive symmetries of machine learning
    arXiv.org2023S2
18. Jacobian-based Causal Discovery with Nonlinear ICA
    Trans. Mach. Learn. Res.2023S2
19. Posterior Annealing: Fast Calibrated Uncertainty for Regression
    arXiv.org2023S2
20. On Data Manifolds Entailed by Structural Causal Models
    International Conference on Machine Learning2023S2
21. Out-of-Variable Generalization
    arXiv.org2023S2
22. Learning Locomotion Skills from MPC in Sensor Space
    Conference on Learning for Dynamics & Control2023S2