Update FrontierOR-v1 release notes

README.md

@@ -34,6 +34,10 @@ paper's OR problem into runnable, verifiably-correct optimization code.
 
 This repository now contains 180 canonical tasks under `tasks/`. The legacy `ostrowski2012` compatibility alias, which duplicated `earl2005`, was removed on 2026-09-06.
 
+## FrontierOR-v1 release notes
+
+FrontierOR-v1 includes an important dataset update that consolidates the canonical task set, Gurobi references, hard-set split metadata, feasibility checkers, and task-instance artifacts. For detailed notes on the fixes and compatibility changes, see [`RELEASE_NOTES.md`](RELEASE_NOTES.md).
+
 ## Layout
 
 | Path | Purpose |

RELEASE_NOTES.md

@@ -1,26 +1,67 @@
-# Release notes: Audited-180 v2
+## v1 Update — 2026-09-27
 
-This revision removes runtime optimization dependencies from the nine
-feasibility checkers that previously invoked Gurobi directly or dynamically:
-`adulyasak2015`, `bertsimas2024`, `cote2018`, `liao2020`, `lozano2017`,
-`rahmaniani2018`, `rahmaniani2022`, `riley2025`, and `taninmis2022`.
+This update consolidates the dataset, reference results, validation logic, and evaluation protocol developed since September 5, 2026. It supersedes earlier v1 snapshots.
 
-All 54 affected reference solutions now carry complete operational variables.
-Seven cases use a sparse `model_variables` map; `lozano2017` retains its full
-leader/follower vectors, and `taninmis2022` retains its named operational maps.
-Six cases also include trusted compressed coefficient snapshots under
-`solverless_models/`. Checkers only perform numeric substitution and do not
-import or invoke Gurobi, SciPy optimizers, HiGHS, CVXPY, or another solver.
+### Canonical dataset
 
-Regression on 2026-09-05 accepted all 54 official pairs, rejected all 54
-objective mutations, and rejected all 54 complete-variable mutations. The
-detailed record is in `audit/SOLVERLESS_CHECKER_AUDIT.json`.
+- FrontierOR now contains **180 canonical tasks**, organized under `tasks/<task_id>/`.
+- The legacy `ostrowski2012` alias was removed because it duplicated `earl2005`.
+- `wang2018` was added to complete the canonical 180-task collection.
+- Paper-level metadata, including optimization direction, is consolidated in:
+  - `metadata/paper_meta_info.json`
+- The canonical 50-task hard subset is published in:
+  - `metadata/splits/hard.json`
+- Its selection criteria, frozen input snapshot, full 180-task ranking, and
+  reproduction script are published in `metadata/splits/` (see
+  `metadata/splits/hard_methodology.md`).
 
-As in v1, acceptance proves feasibility and objective consistency for the
-submitted complete witness. It does not relabel historical incumbents as
-globally optimal. In particular, integer follower/recourse optimality is now a
-solution-quality concern rather than an online checker solve.
 
-## 2026-09-06 alias removal
+### Canonical Gurobi references
 
-Removed the legacy `ostrowski2012` compatibility alias directory, which duplicated `earl2005`. The public dataset now exposes 180 canonical task directories, including `wang2018` in the canonical manifest.
+Canonical Gurobi objectives, benchmark runtimes, time limits, and statuses are now distributed in compact machine-readable files:
+
+- `metadata/gurobi_references.parquet`
+- `metadata/gurobi_references.csv.gz`
+
+These files are the authoritative source for metric computation. They replace the need to reconstruct runtimes from heterogeneous solution fields, incumbent logs, or legacy per-instance CSV files.
+
+
+### Hardened feasibility checkers
+
+The feasibility checkers were systematically audited and strengthened:
+
+- Checkers now validate submitted solutions against trusted instance data and the published solution contract.
+- Objective consistency checks reject missing, inconsistent, or fabricated objective values.
+- Malformed JSON, missing required fields, invalid field types, and variable-domain violations are handled as explicit infeasible results.
+- Capacity, flow, balance, linking, integrality, and task-specific constraints are checked where applicable.
+- Checkers no longer trust fixed problem parameters supplied by a candidate solution.
+- The affected solver-dependent checkers were converted to validate complete witnesses without invoking an optimization solver at evaluation time.
+- Checker execution limits and error handling were made more robust for large solutions.
+
+All published Gurobi reference solutions and evaluated LLM solution artifacts were revalidated using the updated checkers.
+
+
+### Task and instance corrections
+
+We audited the task contracts, instances, reference implementations, and solutions. Notable corrections include:
+
+- `segundo2019`: restored the missing `large_3` and `large_4` instances and updated the corresponding Gurobi artifacts.
+- `wei2020b`: restored the non-empty `large_1` reference solution.
+- `delage2022`: added correct support for the open-ended, closed-ended, and knife-dependent setup variants, including their variant-specific fields and constraints.
+- `bierwirth2017`: corrected the checker to derive operation completion times from trusted instance processing times and removed an unsupported alternative solution schema.
+- `oliveira2020`: regenerated the tiny instance at an appropriate benchmark scale and updated its derived metadata and reference result.
+- `fragkos2021`: corrected an unreliable historical Gurobi reference caused by callback/runtime handling.
+- `furini2016` and `segundo2019`: repaired zero-incumbent fallback handling in the reference Gurobi programs.
+
+Associated reference solutions and feasibility results were regenerated or revalidated where required.
+
+
+### Compatibility notes
+
+Users of earlier FrontierOR snapshots should:
+
+- update task paths to `tasks/<task_id>/`;
+- use `metadata/paper_meta_info.json` for paper metadata;
+- use `metadata/splits/hard.json` for the hard subset;
+- use `metadata/gurobi_references.parquet` or `.csv.gz` for Gurobi references;
+- rerun feasibility checks when relying on results produced by older checker versions.