Commit History
10 days ago
-
Require complete guillotine cutting witnesses for feasibility validation
Verified signatureef42a2e
12 days ago
-
Align initial inventory task contract with reference formulation
Verified signatureClarify priced variable initial stock and enforce its input lower bound. All six published reference solutions and inventory regressions pass.d9b0dbc
13 days ago
-
Shorten forrest2006 release note
Verified signature8e6e2a5 -
Repair degenerate forrest2006 large instances and refresh references
Verified signature8f645ce
15 days ago
-
Correct pettersson2020 objective signs and refresh validated references
Verified signaturee4ab16b
18 days ago
-
Keep JSON as sole paper metadata table and clarify download instructions
Verified signaturec03818a -
Remove model-size metadata section from README
Verified signature5410563 -
Document shared model-size scope once in dataset README
Verified signatureed85d72 -
Replace legacy avg size fields with audited initial-model counts
Verified signatureUpdate only the two metadata files for 180 tasks. Record separate linear, quadratic and general constraints, their sum, and explicit large_instance_1 counting scope. No tasks, figures, scripts or reports uploaded.bec0f18
19 days ago
-
Correct four approved decision-horizon category values
Verified signatureOnly category values for ji2019, kress2019, salemi2022 and wangk2020 change in the two existing metadata files. Preserve HF-current application fields and all other data; no reports, figures or new fields.e2cfe95
25 days ago
-
Update canonical formulation metadata; retain eight core fields only
Verified signatureUpdate only paper_meta_info.json and paper_meta_info_180.csv. Preserve all non-formulation fields; audit provenance stays local.609a71b -
Remove unneeded problem-class metadata fields from JSON and CSV
Verified signatureRemove nine requested fields only; retain the other fields, all 180 task records, and every non-metadata repository file unchanged.3ee5374 -
Retract auxiliary classification artifacts and task documentation changes
Verified signatureRetain only paper_meta_info.json and paper_meta_info_180.csv from the preceding update, as requested. Restore existing task documents and remove newly uploaded audit snapshots, taxonomy, sidecars and figures.2a79d26 -
Retract auxiliary classification artifacts and task documentation changes
Verified signatureRetain only paper_meta_info.json and paper_meta_info_180.csv from the preceding update, as requested. Restore existing task documents and remove newly uploaded audit snapshots, taxonomy, sidecars and figures.96bd751 -
Update structural problem-class labels, audit evidence and variant documentation
Verified signatureIndividually re-review 123 labels and audit 45 source-flagged tasks; preserve instances, executable evaluators and reference results. AI-assisted annotations are not expert-certified.146f33d
-
Sync 19 certified optimal reference labels and fix Gualandi runtime import
Verified signature36c0406 -
Simplify feasibility checker release note
Verified signature3ad7915 -
Fix Amaldi checker numerical overflow and document hardening
Verified signatureValidate finite hyperplane coefficients and use high-precision distances to reject finite overflow attacks while preserving scaling equivalence. All six published references and nine targeted regression checks passed. Update Hardened feasibility checkers in RELEASE_NOTES.md.1f04313
-
Rename hard subset metadata to metadata/hardset
Verified signatureMove the five hard subset metadata files, rename hard.json to hardset.json, and update the reproduction script and dataset documentation. Task IDs and ranking data are unchanged.d71ca3d -
Update dataset card to canonical frontieror namespace
Verified signaturec2b5aa9 -
Normalize Gurobi reference optimality statuses
Verified signatureUpdate the published reference status column to optimal/incumbent based on verified optimality evidence. Objective, runtime, feasibility, time limits, and paths are unchanged. The local proof index is not included in this commit.72c8752
-
Clarify v2 trusted anytime evaluation plan
Verified signature9f5cd54 -
Document trusted anytime evaluation plan
Verified signature85e2f6c
-
Group task corrections by fix principle
Verified signature4ccf751 -
Normalize FrontierOR naming in README
Minwei KongVerified signaturee18177c -
Use FrontierOR branding in dataset README
Minwei KongVerified signature96e8bb6 -
Move project links under README title
Minwei KongVerified signaturea394f68 -
Add project resource links to dataset README
Minwei KongVerified signature55b1d0e -
Add emphasis to FrontierOR-v1 README heading
Minwei KongVerified signature48ead86 -
Polish FrontierOR-v1 release notes wording
Minwei KongVerified signaturec34c9ed -
Refresh README for FrontierOR-v1 update
Minwei KongVerified signature6d5cab6 -
Update FrontierOR-v1 release notes
Minwei KongVerified signatureaa2a8a3 -
Remove root case catalog
Minwei KongVerified signatureRemove case_catalog_180.json from the dataset root after standardizing all tasks to one tiny and five large official pairs.20904d9 -
Remove redundant second tiny instances
Minwei KongVerified signatureRemove tiny_instance_2 reference artifacts from 17 tasks so every canonical task has one tiny and five large official pairs. Refresh case_catalog_180.json official_pair_count values to 6 for all 180 tasks.0721cee -
Clean root solverless utilities and refresh case catalog
Minwei KongVerified signatureRemove root-level solverless utility scripts that are not required by task-local checkers, and refresh case_catalog_180.json official_pair_count values from current task folders.53c1722 -
Clean obsolete root release files
Minwei KongVerified signatureRemove obsolete UPSTREAM_README, BUILD_INFO, and release manifest files from the dataset root. Keep lightweight case catalog and metadata splits as the public indices.ec622c6 -
Keep hard selection documentation focused on published rule
Minwei KongVerified signatured9ee6b5 -
Explain two-stage hard-set selection
Minwei KongVerified signature5a9d46c -
Cap hard-set median runtime at solve budget
Minwei KongVerified signature9378a1b -
Align hard split IDs with published ranking
Minwei KongVerified signatured7fdd31 -
Tighten hard split methodology wording
Minwei KongVerified signature3ebad2d -
Fix hard split reproduction from HF snapshot
Minwei KongVerified signaturea490828 -
Document and reproduce frozen hard-50 selection
Minwei KongVerified signature61389b7
-
Add consolidated Gurobi reference tables
Minwei KongVerified signature66b12c6 -
Standardize wang2018 runtime metadata
Minwei KongVerified signaturePreserve end-to-end runtime for evaluator compatibility while retaining Gurobi model.Runtime as solve_time_seconds for diagnostics.1dffc7e
-
Update fragkos2021 and oliveira2020 references
Minwei KongVerified signatureRefresh the audited fragkos2021 tiny Gurobi certificate. Replace the oversized oliveira2020 tiny instance with the reproducible 8-job instance and proven reference, and correct derived metadata on its five large instances.91ca9cc -
Add wang2018 to canonical manifest
Minwei KongVerified signatureAdd wang2018 to canonical case indices and upload gurobi_code.py with real _GUROBI_CODE_START_TIME initialization.e973f26
-
Sync remaining audited dataset files
Minwei KongVerified signatureSynchronize 33 locally audited task, schema, checker, Gurobi, and hard-split files. Release notes and version tag will be added separately.e5ab271
-
Add wang2018 as the 180th canonical task
Minwei KongVerified signaturePublish the selected 180th task and update canonical metadata and dataset documentation to the 180-task release.cbf9c08
-
Fix bierwirth2017 trusted processing times and canonicalize delage2022 update
Minwei KongVerified signature1063d09