Search results

Solve-allowed verification: one round-trip check (solve_mismatch) that subsumes all violation types

enhancementNew feature or requestNew feature or request

Status: Open.

#16 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 17, 2026

Cheaply-verifiable rule for a buggy reduction, generalized to optimization (no solving)

enhancementNew feature or requestNew feature or request

Status: Open.

#15 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 17, 2026

Easier ways to identify a buggy rule

Status: Open.

#14 In CodingThrust/problem-reductions-benchmark;

· isPANN opened on Jun 17, 2026

M4.1d — Observability: per-session cost and failure report

Status: Open.

#13 In CodingThrust/problem-reductions-benchmark;

· Ferrari-72 opened on Jun 16, 2026

M4.1c — Session reproducibility archive

Status: Open.

#12 In CodingThrust/problem-reductions-benchmark;

· Ferrari-72 opened on Jun 16, 2026

M4.1b — Calibration tests in CI

Status: Open.

#11 In CodingThrust/problem-reductions-benchmark;

· Ferrari-72 opened on Jun 16, 2026

M4.1a — README and contributor guide

Status: Open.

#10 In CodingThrust/problem-reductions-benchmark;

· Ferrari-72 opened on Jun 16, 2026

📖 Read me first — benchmark overview & glossary

Status: Open.

#9 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 10, 2026

M4.1 — [Tracking] Harden and release the benchmark

Status: Open.

#8 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 10, 2026

·  

M4: Hardening & release

M3.2 — CI: rebuild the index and publish the site, schema-gated

Status: Open.

#7 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 10, 2026

·  

M3: Public website

M3.1 — Build the public leaderboard: dual metrics + certificate drilldown + trajectory

Status: Open.

#6 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 10, 2026

·  

M3: Public website

M2.2 — Harden the verifier (robust equality) and add the novelty filter

Status: Open.

#5 In CodingThrust/problem-reductions-benchmark;

· GiggleLiu opened on Jun 10, 2026

·  

M2: Fair multi-model + trustworthy judgment