- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
Self-Curriculum orders verified, self-generated solutions by the learner's length-normalized negative log-likelihood. It periodically reranks unused trajectories and gives less likely correct solutions higher rewards. Experiments cover mathematical reasoning, with additional model scaling and non-mathematical evaluations.
Critical strengths
- The controls are unusually helpful: student-generated pools are shared across difficulty signals, and filtering, ordering, reranking, and reward shaping are examined separately.
- Improvements survive independent training seeds and extend to 14B models. The reported overhead and expanded-rollout GRPO comparison make the practical case reasonably convincing.
Critical weaknesses
- The ranking remains sensitive to wording and notation despite length normalization. Figure 2 tracks likelihood on trajectories used in training, so it offers limited evidence that the score identifies a useful learning progression beyond fitting those particular strings.
- The custom NuminaMath split is described mainly through source counts. Duplicate or related contest problems are not explicitly grouped before splitting, and the AMC/AIME test subset has only 42 questions. The harder-math generalization evidence is therefore less secure than the main MATH results.
Questions for authors
- How stable are difficulty rankings under equivalent rewrites of the same correct solution?
- What deduplication or problem-family grouping was applied before constructing the NuminaMath train/test split?