Logged bandit feedback trains calibrated decision heads on Qwen3
CAPS beats confirmed-answer fine-tuning by 2.6 accuracy points on Banking77 and cuts ECE by 0.053.
38
CAPS beats confirmed-answer fine-tuning by 2.6 accuracy points on Banking77 and cuts ECE by 0.053.
SCAPO uses semifactual token stability to beat GRPO on AIME for two Qwen3 base models.
SCAPO adds semifactual token stability to GRPO, lifting Qwen3 AIME accuracy by up to 5.63 points.
LastOPD briefly aligns only the last-layer latent state when distilling Qwen3 into Qwen3-1.7B, beating token-only OPD on MATH-500.
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.
Today 112 readers · 115 views · 24 reads · all time 199K views