Unverified50% confidenceFactExact time
DPO之后社区涌现出SimPO、ORPO、KTO等一系列变体
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
FTPO: A Training-Level Solution to AI Repetition Loops
hackernewshackernews7/7/2026
Related Claims
UnverifiedPreference alignment with the DPO algorithm requires preparing both positive examples (preferred responses) and negative examples (dispreferred responses).53% similarVerifiedDPO (Direct Preference Optimization) is an alignment method proposed in 2023 that can skip reward model training and directly optimize the policy model from human preference data.51% similarUnverified微调后的模型可通过DPO（直接偏好优化）进行对齐，确保输出符合业务安全规范50% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/396333API
curl https://kongchang.com/api/v1/knowledge/claims/396333MCP
get_claim(id=396333)