cs.CLNov 10, 2026

How Post-Training Influences Verbalized Algorithmic Computation in Language Models

Authors: Niall Dalton

Organizations: University College London

Abstract

Post-training improves language-model reasoning, but how it changes the use of internal information and verbalized chain of thought (CoT) remains unclear. We study successive SFT, DPO, and RLVR checkpoints of Olmo-3-7B-Instruct on multi-digit addition, using carry values as known intermediate state. Probes recover carry information from internal activations with similar accuracy in SFT and RLVR, although manipulating this information does not reliably control model outputs. In contrast, the usefulness of verbalized reasoning differs: CoTs produced by RLVR yield higher accuracy across receiving checkpoints than those produced by SFT or DPO. We also show that carry-related statements receive higher importance scores after RLVR than matched random controls. DPO shows similar sensitivity to carry-related statements but without comparable accuracy gains. These results suggest that post-training changes the usefulness of written reasoning and sensitivity to its content in meaningful ways, although greater sensitivity does not necessarily coincide with better problem performance.

Figures & tables

Appendix figures & tables29 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Post-Training Shifts Confidence: A Three-Stage Analysis of How SFT, RL, and OPD Shape Pre-, Intra-, and Post-CoT Calibration

    Jul 15, 2026Shuhao Li, Guodong Du, Anhao Zhao +3Confidence Estimation in Language ModelsLanguage Model Calibration

  2. What Does Post-Training Change in Multilingual Reasoning?

    Sep 29, 2026Hongyang Li, Xiao Li, Caesar Wu +2Multilingual Language Model EvaluationCross-Lingual Reasoning in Language Models