Direct Preference Optimization

Technique

First seen 7/16/2026

Last seen 7/16/2026

Evidence 1 chunks

NEIGHBORHOOD

No graph connections found for this entity yet. It may appear in future ingestion runs.

2 connections

GoldenFuzz ← uses 100% 1e

GoldenFuzz applies Direct Preference Optimization (instantiated as SimPO) to refine instruction generation policy.

SimPO ← extends 90% 1e

SimPO is a variant/extension of Direct Preference Optimization used for policy updates in GoldenFuzz.