DPO
by Stanford · open-source · Last verified 2026-04-24
Direct Preference Optimization — alignment without a separate reward model.
https://arxiv.org/abs/2305.18290 ↗C
C—Below Average
Adoption: C+Quality: B+Freshness: ACitations: CEngagement: F
Specifications
- License
- Open Source
- Pricing
- open-source
- Capabilities
- Integrations
- Use Cases
- API Available
- No
- SDK Languages
- Tags
- safety, alignment, training, fine-tuning
- Added
- 2026-04-24
- Completeness
- 60%
Index Score
44Adoption
50
Quality
70
Freshness
80
Citations
40
Engagement
0