← Back to Publications

StateShift: State-Matched Evaluation of Error Recovery in Neural Reasoning

Sham Satish Thakare

Abstract

Formulates state-matched continuation evaluation protocols under verifier perturbations on neural reasoning rollouts. Evaluates structural covariate controls and error recovery dynamics across language-model reasoning trajectories.

Artifacts & Code

BibTeX

@article{thakare2026stateshift,
  author = {Thakare, Sham Satish},
  title  = {StateShift: Tracking State-Dependent Reasoning Recovery Across Post-Training},
  year   = {2026},
  note   = {Preprint / Technical Report},
  url    = {https://github.com/shamddd/stateshift}
}