CRITICAL BUG FIX: Clamp reward on ALL early return paths - prevents returning 0.0 6138889 Nanda Kumar Kondreddy commited on Apr 11
Fix: Enforce strict open interval (0.01, 0.99) for all grader rewards - validator compliance 8e7ae76 Nanda Kumar Kondreddy commited on Apr 11
Phase 2: Benchmark Quality Upgrades - Progressive graders, realistic multi-bug tasks, ground_truth exposure b43aff5 Nanda Kumar Kondreddy commited on Apr 11