https://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/dpo-variants-ipo-kto-orpo-and-cdpo2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/identity-preference-optimization-ipo2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/kahneman-tversky-optimization-kto2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/odds-ratio-preference-optimization-orpo2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/conservative-dpo-cdpo2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/worked-example-computing-losses-step-by-step2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/implementation-comparison2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/comparing-alignment-methods2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/practical-guidelines2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/limitations-and-open-problems2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/summary2026-01-03T10:00:00.000Zhttps://mbrenndoerfer.com/writing/dpo-variants-ipo-kto-orpo-cdpo-llm-alignment/quiz2026-01-03T10:00:00.000Z