[KongchangAI]
Unverified50% confidenceTradeoffExact time

The cost of DPO is that it places high demands on the distribution coverage of preference data, and it is difficult to flexibly introduce online exploration as PPO can.

1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen

Sources

Related Entities

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/1025165
API
curl https://kongchang.com/api/v1/knowledge/claims/1025165
MCP
get_claim(id=1025165)