method · first seen 5 Oct 2026
rejection-sampling fine-tuning
Also: RFT, rejection-sampled trajectories
Training a model on its own sampled outputs that pass a correctness check, keeping only successful samples.
Mentions per week
PapersCodeDiscussion
Earliest sources
- What pass@k Cannot Measure: Evaluating Diversity and Capability Retention after Post-TrainingPapers · 5 Oct 2026 · Subham Rath, Raj Dandekar, Rajat Dandekar, Sreedath Panat
- CodeWalk: Generating Coding Benchmarks by Walking a Problem GraphPapers · 11 Oct 2026 · Jiaxin Fang, Runyuan He, Arnav Gupta, Sahil Kumawat, Huanzhi Mao, Alvin Cheung
- CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow?Papers · 18 Oct 2026 · Jiaqi Su, Guibin Zhang, Jie-Jing Shao, Cong Pang, Tiankuo Yao, Ivor Tsang, Haiya…
- Is it RL or Data? A Controlled Comparison of GRPO and Rejection-Sampling Fine-TuningPapers · 19 Oct 2026 · Lakshya Chaudhry
- Concise-9B: Removing the Slack from a Reasoning Model's ThinkingPapers · 1 Nov 2026 · Alexander Boesgaard Lorup
Latest sources
- Concise-9B: Removing the Slack from a Reasoning Model's ThinkingPapers · 1 Nov 2026 · Alexander Boesgaard Lorup
- Is it RL or Data? A Controlled Comparison of GRPO and Rejection-Sampling Fine-TuningPapers · 19 Oct 2026 · Lakshya Chaudhry
- CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow?Papers · 18 Oct 2026 · Jiaqi Su, Guibin Zhang, Jie-Jing Shao, Cong Pang, Tiankuo Yao, Ivor Tsang, Haiya…
- CodeWalk: Generating Coding Benchmarks by Walking a Problem GraphPapers · 11 Oct 2026 · Jiaxin Fang, Runyuan He, Arnav Gupta, Sahil Kumawat, Huanzhi Mao, Alvin Cheung
- What pass@k Cannot Measure: Evaluating Diversity and Capability Retention after Post-TrainingPapers · 5 Oct 2026 · Subham Rath, Raj Dandekar, Rajat Dandekar, Sreedath Panat