Second Thoughts are Best Learning to Re Align With Human Values from Text Edits Appendix

A.1 Detailed Re-alignment Task Formulation and Training Setup In Figure A1, we show the procedure for converting the data samples in the alignment datasets into training data of AEM (negative samples used in AIL are generated similarly). In DP-inferred chain-of-edits (CoEs), we use a few special tokens to mark the editing operations (with their position and content). Then our decipher module will translate these special tokens into natural language. As the final step, we add a special token [SEP] between Context + Source and the ground truth Chain-of-Edits (CoEs) and Target, as a boundary signal similar to the settings in text-to-text training. We also augment the data by using different sets of costs for the editing operations (as discussed in Section 3.2, and footnote 3).