@article{probebaseddataattributiondiscoveringandm, title = {Probe-Based Data Attribution: Discovering and Mitigating Undesirable Behaviors in LLM Post-Training}, author = {Frank Xiao and Santiago Aranguri}, year = {2026}, eprint = {2602.11079}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2602.11079}, }