e-process

Here are 5 public repositories matching this topic...

kadubon / audit-closed-ai-scientist

Benchmark for statistically valid AI scientist systems, using audit-closed protocols, transparency logs, and sequential inference to prevent false discoveries in autonomous research agents.

Updated Aug 28, 2026
Python

dthinkr / bet-on-features

Star

Find feature-specific failures in quantile forecasts as outcomes arrive.

calibration online-learning quantile-forecasting e-process

Updated Jul 19, 2026
Python

paulchum / verdict-bench

Star

Certified branch/tool gating evidence harness for public agent workloads

reproducible-research llm-agents agent-evaluation e-process tau-bench

Updated Jul 31, 2026
Python

CJX0712 / certiscale

Star

CertiScale 衡证 · 可证明的推理时计算规模化：CITE × 验证器校准 e-process × 共享预算分配。Anytime-valid Type-I 错误可证，FDR 较 CITE 降 21-43×。作者：晨星

artificial-intelligence uncertainty-quantification self-consistency conformal-prediction sequential-testing budget-allocation llm test-time-compute e-process anytime-valid

Updated Oct 12, 2026
Python

MG-05 / E-Process-Authorized-Thompson-Sampling

Star

Official code and reproducibility materials for e-ATS: e-process-authorized Thompson sampling for non-stationary bandits. NeurIPS 2026 E-Values Workshop.

thompson-sampling multi-armed-bandits e-process

Updated Oct 28, 2026
Python

Add this topic to your repo

To associate your repository with the e-process topic, visit your repo's landing page and select "manage topics."

Learn more