LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling Paper • 2606.18023 • Published Jun 16 • 210
MIRA: Mid-training Rubric Anchoring for Source-Aware Data Selection Paper • 2605.30288 • Published May 29 • 23
MIRA Collection Group-specific quality scorers from MIRA for mid-training data selection. • 12 items • Updated May 27 • 2