R0040/2026-04-01/Q001/S04/R04¶
Original SPIN paper on self-play fine-tuning.
Summary¶
| Field | Value |
|---|---|
| Title | Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models |
| URL | https://arxiv.org/abs/2401.01335 |
| Date accessed | 2026-04-01 |
| Publication date | 2024-01-02 (ICML 2024) |
| Author(s) | Zixiang Chen et al. (UCLA) |
| Publication | ICML 2024 |
Selection Decision¶
Included in evidence base: Yes
Rationale: Original peer-reviewed paper introducing SPIN. Demonstrates self-play as a viable alignment paradigm.