Publish NL optimization instantiation metrics v1

Initial scientific release payload for reduced text-free derived metrics. No raw NLP4LP content or private audit artifacts included.

CHECKSUMS.sha256

@@ -0,0 +1,17 @@
+01c0753f14f336d6e0fcbc354fed4eb8fbb792a903159379eb954c0ef9d8a372  CITATION.cff
+e8ef6f98306d0ec0909700ee101b7b12031cdcbaeec67a03e3b9cb04dbf6c6a0  LICENSE
+1f25c59b5627dcf94ccb5a8d0181a3e236b5fbe0279af9295ec4a71c3b2a18a8  README.md
+0a66e012b9e0e6ae48e4eeaea16ef73b86a642a7f42c64a370b80a814bc921af  data/method_summary/train.parquet
+953b211f09e380166cd8cb5c3da3207ae5774ea893404858fca3f7d81bc09434  data/patch_case_metrics/train.parquet
+1955b49d85059ecb812ee266c400d4b0cfe703e42507d81ed2783fed42c0f1ec  data/patch_transitions/train.parquet
+1703aeda13050a7324e9e4979947f11890c41afeaec80693194aaf09cae4469e  data/per_query_metrics/train.parquet
+8bfda830c1fed16ca826c78a76e3281a90a0d296445cc96c933af1f53ebadaef  data/strict_readiness_summary/train.parquet
+00e4fbb19d475bc2bbfc57807c5aa2cefd18c3a652766247d5da283784f5c520  metadata/final_validation_report.json
+731cea668a7587abda2915fa4b87ed887aee63f4a9ed5a3a029ee7e4870ed0eb  metadata/provenance_summary.json
+1c9c7c1e681e00f44c92123333bcf5a236ff7975555fea0fb17712a2a6506b06  metadata/reconstruction_risk_report.json
+7f201444540552e12731a1b044c5eb48b07817367da554282f76182a813e8f4a  metadata/release_manifest.json
+b697664d1accb2adbcb9234f463e8763c4c72519f14a68b60fb548359b593f20  metadata/rights_scope.json
+a3dd4b13ebb8f47ac7ca2679dd563de15964d5d1b0173faf345a028f47eca5e9  metadata/schema.json
+50198da5225b674f198d8e4da3ef0d6e47a06d26e08d2e043e76c7ff267f36e6  metadata/security_scan_report.json
+e65a741e3ade7681d2d7e0c6a568d6560bfd7fa5c9da58566c911908c877e7c8  metadata/statistics.json
+c6e1b9d5e39957ce3e84f07de02ecb848d5fcc2796258d2c09c3eddd045428af  metadata/upstream_attribution.md

CITATION.cff

@@ -0,0 +1,33 @@
+cff-version: 1.2.0
+title: Natural-Language Optimization Instantiation Metrics
+message: Cite this dataset when using the released rows or metrics. Cite the associated paper for methodology/results and cite NLP4LP/OptiMUS for upstream benchmark provenance.
+type: dataset
+authors:
+  - family-names: Vahidi
+    given-names: Soroush
+repository-code: https://github.com/SoroushVahidi/combinatorial-opt-agent
+url: https://huggingface.co/datasets/SoroushVahidi/nl-optimization-instantiation-metrics
+date-released: 2026-08-14
+version: v1.0.0
+license: CC-BY-NC-SA-4.0
+references:
+  - type: preprint
+    title: Retrieval-Assisted Instantiation of Natural-Language Optimization Problems
+    authors:
+      - family-names: Vahidi
+        given-names: Soroush
+    year: 2026
+    doi: 10.2139/ssrn.6434886
+  - type: conference-paper
+    title: 'OptiMUS: Scalable Optimization Modeling with (MI)LP Solvers and Large Language Models'
+    authors:
+      - family-names: Ahmaditeshnizi
+        given-names: Ali
+      - family-names: Gao
+        given-names: Wenzhi
+      - family-names: Udell
+        given-names: Madeleine
+    year: 2024
+    conference:
+      name: Proceedings of the 41st International Conference on Machine Learning
+    url: https://proceedings.mlr.press/v235/ahmaditeshnizi24a.html

LICENSE

@@ -0,0 +1,2 @@
+Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
+https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode

README.md

@@ -0,0 +1,198 @@
+---
+license: cc-by-nc-sa-4.0
+task_categories:
+- tabular-classification
+- tabular-regression
+language:
+- en
+tags:
+- optimization
+- optimization-modeling
+- natural-language-optimization
+- evaluation
+- metrics
+- retrieval
+- grounding
+- lp
+- milp
+size_categories:
+- n<1K
+configs:
+- config_name: per_query_metrics
+  data_files:
+  - split: train
+    path: data/per_query_metrics/train.parquet
+- config_name: method_summary
+  data_files:
+  - split: train
+    path: data/method_summary/train.parquet
+- config_name: strict_readiness_summary
+  data_files:
+  - split: train
+    path: data/strict_readiness_summary/train.parquet
+- config_name: patch_transitions
+  data_files:
+  - split: train
+    path: data/patch_transitions/train.parquet
+- config_name: patch_case_metrics
+  data_files:
+  - split: train
+    path: data/patch_case_metrics/train.parquet
+---
+
+# Natural-Language Optimization Instantiation Metrics
+
+This dataset is a text-free metrics release for a retrieval-assisted natural-language optimization instantiation pipeline. It contains per-example and aggregate evaluation outcomes for a frozen pipeline evaluated on the NLP4LP benchmark.
+
+Soroush Vahidi created and releases the evaluation metrics and derived outcomes in this repository. NLP4LP is a third-party benchmark/resource maintained by the Udell Lab. Soroush Vahidi does not own NLP4LP or OptiMUS.
+
+Raw NLP4LP problem text, gold formulations, gold structures, source IDs, prompts, completions, external example text, provider raw outputs, and local/private metadata are not included. Public rows use project-local opaque identifiers. Users who need the source benchmark must obtain NLP4LP directly from its canonical source under its own access and licensing terms: https://huggingface.co/datasets/udell-lab/NLP4LP
+
+## What Is Being Evaluated?
+
+Many optimization problems are first written in natural language and later translated into mathematical models. This release evaluates a narrower intermediate task: natural-language optimization instantiation. Given a natural-language problem description, the evaluated system retrieves a compatible optimization schema from a fixed catalog and then fills scalar parameter slots from numeric evidence.
+
+Schema retrieval means selecting the top-ranked optimization schema for a problem description. Instantiation readiness measures whether the retrieved and filled schema has enough correctly typed scalar information to be useful for downstream modeling.
+
+The frozen method represented here is TF-IDF schema retrieval plus typed greedy scalar grounding with deterministic multiplicative ratio-word extraction. The patch adds deterministic extraction for expressions such as multiplicative ratio words, while leaving schema retrieval, slot choice, and evaluation thresholds unchanged.
+
+## Configs
+
+### `per_query_metrics`
+
+One row represents one benchmark query evaluated by the frozen method. It does not include the native benchmark ID or text.
+
+Fields:
+
+- `example_uid`: project-local opaque identifier derived from the native source ID using a private keyed HMAC; the key and mapping are not released.
+- `variant`: benchmark/query variant label from the source evaluation output. In this v1 release it is `orig`.
+- `method`: method identifier. This is the source `baseline` column renamed for clarity; in this v1 release it is `tfidf`.
+- `schema_hit`: whether the retrieved top-1 schema matched the gold schema.
+- `ordinary_ready`: `param_coverage >= 0.8 AND type_match >= 0.8`.
+- `strict_ready`: `schema_hit AND param_coverage >= 0.8 AND type_match >= 0.8`.
+- `n_expected_scalar`: structural count of expected scalar slots used for evaluation. It is a small integer count and does not include slot names, variable names, formulation text, or problem text.
+- `n_filled`: number of scalar slots filled by the evaluated method.
+- `param_coverage`: scalar-slot coverage metric.
+- `type_match`: scalar-slot type compatibility metric.
+- `exact5`: exact-value metric at 5% tolerance where comparable; null means no comparable exact-value denominator for that row.
+- `exact20`: exact-value metric at 20% tolerance where comparable; null means no comparable exact-value denominator for that row.
+- `key_overlap`: schema/key overlap metric.
+
+### `method_summary`
+
+Aggregate metrics for the pre-patch and final patched method variants. Counts are integer counts; rates and means are floating-point metrics.
+
+### `strict_readiness_summary`
+
+Small aggregate table for strict-readiness counts and rate changes.
+
+### `patch_transitions`
+
+Transition counts comparing pre-patch and patched outcomes. ID lists from the source transition file are deliberately excluded.
+
+### `patch_case_metrics`
+
+Redacted per-case patch metrics for the changed-query audit. This config excludes source sentences, native query IDs, extracted mention strings, affected slot names, and multiplier expression text. It keeps only opaque IDs, change-status categories, readiness booleans, and numerical before/after metrics.
+
+## Current Canonical Results
+
+The final patched method has:
+
+- schema R@1: `0.909366`
+- ordinary instantiation-ready: `265 / 331 = 0.800604`
+- strict instantiation-ready: `255 / 331 = 0.770393`
+- strict-ready patch gain: `+8`
+- strict-ready losses: `0`
+
+These are point-in-time results for the frozen method state represented by the source repository at commit `6bb75a4c4bed02c458ac30b4af206a2802fce095`.
+
+## Limitations
+
+This is a benchmark-specific metrics release over NLP4LP `orig` evaluation queries. It does not contain the source benchmark examples, so users cannot inspect or resolve individual examples from this dataset alone. Users needing problem text or gold structures must obtain NLP4LP from the canonical upstream source.
+
+The dataset includes remaining non-ready cases, schema retrieval failures, and type/coverage failures. Null `exact5` and `exact20` values indicate rows without a comparable exact-value denominator, not successful or failed exact matching.
+
+The metrics are derived from one frozen pipeline/version. They should not be treated as claims about all natural-language optimization systems or all optimization-modeling benchmarks.
+
+## How This Dataset Differs from Existing Resources
+
+NLP4LP provides benchmark problem resources. OptiMUS provides optimization-modeling systems and papers around NLP4LP. NL4Opt, ORLM, and PaMOP are related natural-language optimization or optimization-modeling resources. This release instead provides a text-free, downloadable metrics artifact: project-generated per-example evaluation outcomes, readiness indicators, and aggregate method metrics for a specific retrieval-assisted instantiation pipeline.
+
+In the resources examined for this release, we did not identify an equivalent downloadable metrics-only artifact with this exact evaluation representation.
+
+## Relationship to Existing SoroushVahidi Datasets
+
+This dataset is distinct from the existing original research datasets under `SoroushVahidi`:
+
+- `lafc-evict`: cache-eviction/counterfactual supervision data.
+- `module-intervention-credit`: LLM-serving intervention credit simulation results.
+- `consistency-aware-judgments`: consistency-aware ranking/judgment data.
+- `frontier-allocation-metrics`: budgeted inference/frontier allocation metrics.
+- `mwfas-heuristic-metrics`: minimum-weighted feedback arc set heuristic metrics.
+- `ranking-fas-results`: ranking by feedback arc set results.
+
+This release concerns natural-language optimization problem instantiation and retrieval/grounding readiness. It does not supersede any existing dataset.
+
+## Ownership, Provenance, and License
+
+Soroush Vahidi is the dataset author for this derived metrics release.
+
+The released derived metrics, annotations, and release documentation are licensed under CC BY-NC-SA 4.0. This license does not relicense NLP4LP itself. NLP4LP remains subject to its own upstream terms, access controls, and license.
+
+Observed upstream NLP4LP metadata at build time:
+
+- Canonical source: https://huggingface.co/datasets/udell-lab/NLP4LP
+- Maintainer: Udell Lab
+- Hugging Face license tag: `cc-by-nc-sa-4.0`
+- Access: gated
+
+## Associated Paper
+
+Direct associated paper:
+
+Soroush Vahidi. Retrieval-Assisted Instantiation of Natural-Language Optimization Problems. Preprint records: SSRN DOI `10.2139/ssrn.6434886`; Research Square DOI `10.21203/rs.3.rs-10463941/v1`.
+
+## Citation Policy
+
+Cite this dataset when using the released rows or metrics. Cite the associated paper when using or discussing the methodology or scientific results. Cite both when using the released data and relying materially on the associated methodology/results. Cite NLP4LP/OptiMUS for upstream benchmark provenance.
+
+Dataset citation:
+
+```bibtex
+@dataset{vahidi_nl_optimization_instantiation_metrics_2026,
+  title = {Natural-Language Optimization Instantiation Metrics},
+  author = {Vahidi, Soroush},
+  year = {2026},
+  publisher = {Hugging Face},
+  howpublished = {\url{https://huggingface.co/datasets/SoroushVahidi/nl-optimization-instantiation-metrics}},
+  note = {Derived metrics release; no dataset DOI assigned}
+}
+```
+
+Associated paper citation:
+
+```bibtex
+@misc{vahidi_retrieval_assisted_instantiation_2026,
+  title = {Retrieval-Assisted Instantiation of Natural-Language Optimization Problems},
+  author = {Vahidi, Soroush},
+  year = {2026},
+  doi = {10.2139/ssrn.6434886},
+  note = {Preprint; Research Square DOI: 10.21203/rs.3.rs-10463941/v1}
+}
+```
+
+Upstream benchmark citation:
+
+```bibtex
+@inproceedings{ahmaditeshnizi2024optimus,
+  title = {{O}pti{MUS}: Scalable Optimization Modeling with ({MI}){LP} Solvers and Large Language Models},
+  author = {Ahmaditeshnizi, Ali and Gao, Wenzhi and Udell, Madeleine},
+  booktitle = {Proceedings of the 41st International Conference on Machine Learning},
+  pages = {577--596},
+  year = {2024},
+  volume = {235},
+  series = {Proceedings of Machine Learning Research},
+  publisher = {PMLR},
+  url = {https://proceedings.mlr.press/v235/ahmaditeshnizi24a.html}
+}
+```

data/method_summary/train.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a66e012b9e0e6ae48e4eeaea16ef73b86a642a7f42c64a370b80a814bc921af
+size 9380

data/patch_case_metrics/train.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:953b211f09e380166cd8cb5c3da3207ae5774ea893404858fca3f7d81bc09434
+size 11718

data/patch_transitions/train.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1955b49d85059ecb812ee266c400d4b0cfe703e42507d81ed2783fed42c0f1ec
+size 5223

data/per_query_metrics/train.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1703aeda13050a7324e9e4979947f11890c41afeaec80693194aaf09cae4469e
+size 14469

data/strict_readiness_summary/train.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bfda830c1fed16ca826c78a76e3281a90a0d296445cc96c933af1f53ebadaef
+size 2400

metadata/final_validation_report.json

@@ -0,0 +1,28 @@
+{
+  "all_null_columns": [],
+  "all_null_columns_ok": true,
+  "dataset_server_readiness_expected": true,
+  "expected_row_counts_match": true,
+  "flat_schema_no_nested_or_binary": true,
+  "logical_predicates_are_bool": true,
+  "notes": [
+    "No upload performed.",
+    "No Hugging Face repository created.",
+    "No raw NLP4LP content included.",
+    "No canonical source outputs modified."
+  ],
+  "overall_status": "READY_FOR_HUGGING_FACE_PUBLICATION",
+  "private_audit_material_not_in_public_staging": true,
+  "reconstruction_risk_status": "PASS",
+  "row_counts": {
+    "method_summary": 2,
+    "patch_case_metrics": 36,
+    "patch_transitions": 5,
+    "per_query_metrics": 331,
+    "strict_readiness_summary": 3
+  },
+  "security_scan_status": "PASS",
+  "source_hashes_match_after_build": true,
+  "source_head_after": "6bb75a4c4bed02c458ac30b4af206a2802fce095",
+  "source_worktree_clean_after": true
+}

metadata/provenance_summary.json

@@ -0,0 +1,18 @@
+{
+  "identifier_policy": "Public rows use project-local opaque HMAC-derived identifiers. The key and native-ID mapping are private audit artifacts and are not part of staging.",
+  "not_included": [
+    "raw NLP4LP problem text",
+    "gold formulations",
+    "gold structures",
+    "native source IDs",
+    "prompts",
+    "completions",
+    "provider raw outputs",
+    "local paths/private metadata"
+  ],
+  "released_content_owner": "Soroush Vahidi created and releases the evaluation metrics and derived outcomes in this repository.",
+  "upstream_benchmark": "NLP4LP, a third-party benchmark/resource maintained by the Udell Lab.",
+  "upstream_dataset_url": "https://huggingface.co/datasets/udell-lab/NLP4LP",
+  "upstream_hf_sha_observed": "af17e5178519a786ec0dd686d57607bc813c1af9",
+  "upstream_license_observed": "cc-by-nc-sa-4.0"
+}

metadata/reconstruction_risk_report.json

@@ -0,0 +1,13 @@
+{
+  "can_reconstruct_gold_formulation": false,
+  "can_reconstruct_gold_or_retrieved_document_identifiers": false,
+  "can_reconstruct_mention_strings": false,
+  "can_reconstruct_native_ids": false,
+  "can_reconstruct_problem_text": false,
+  "can_reconstruct_slot_names": false,
+  "can_reconstruct_source_sentences": false,
+  "forbidden_private_or_native_terms_in_public_docs": [],
+  "forbidden_terms_in_parquet_payload": [],
+  "notes": "The public payload contains opaque HMAC-derived IDs, method labels, booleans, counts, and numeric metrics only. The private HMAC key and native-ID mapping are not in staging.",
+  "status": "PASS"
+}

metadata/release_manifest.json

@@ -0,0 +1,48 @@
+{
+  "configs": {
+    "method_summary": {
+      "path": "data/method_summary/train.parquet",
+      "rows": 2
+    },
+    "patch_case_metrics": {
+      "path": "data/patch_case_metrics/train.parquet",
+      "rows": 36
+    },
+    "patch_transitions": {
+      "path": "data/patch_transitions/train.parquet",
+      "rows": 5
+    },
+    "per_query_metrics": {
+      "path": "data/per_query_metrics/train.parquet",
+      "rows": 331
+    },
+    "strict_readiness_summary": {
+      "path": "data/strict_readiness_summary/train.parquet",
+      "rows": 3
+    }
+  },
+  "created_utc": "2026-08-14T02:31:05.073919+00:00",
+  "date_released_planned": "2026-08-14",
+  "included_optional_patch_case_metrics": true,
+  "license": "CC BY-NC-SA 4.0",
+  "public_files_minimum": [
+    "README.md",
+    "LICENSE",
+    "CITATION.cff",
+    "CHECKSUMS.sha256",
+    "data/",
+    "metadata/"
+  ],
+  "repo_id_intended": "SoroushVahidi/nl-optimization-instantiation-metrics",
+  "source_branch": "main",
+  "source_hashes_recorded_private": true,
+  "source_head": "6bb75a4c4bed02c458ac30b4af206a2802fce095",
+  "source_remotes": [
+    "origin\thttps://github.com/SoroushVahidi/combinatorial-opt-agent.git (fetch)",
+    "origin\thttps://github.com/SoroushVahidi/combinatorial-opt-agent.git (push)"
+  ],
+  "source_repository": "SoroushVahidi/combinatorial-opt-agent",
+  "source_repository_path_redacted": true,
+  "source_worktree_clean": true,
+  "version": "v1.0.0"
+}

metadata/rights_scope.json

@@ -0,0 +1,8 @@
+{
+  "does_not_relicense_nlp4lp": true,
+  "license_for_released_metrics": "CC BY-NC-SA 4.0",
+  "native_ids_included": false,
+  "nlp4lp_must_be_obtained_separately": true,
+  "raw_upstream_text_included": false,
+  "reconstruction_risk_assessment": "No meaningful reconstruction path from public payload alone for NLP4LP text, gold formulations, slot names, mention strings, source sentences, or native IDs."
+}

metadata/schema.json

@@ -0,0 +1,272 @@
+{
+  "method_summary": [
+    {
+      "arrow_type": "large_string",
+      "name": "method",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "n",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "schema_r1",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "schema_correct_count",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "coverage_mean",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "type_match_mean",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "ordinary_instantiation_ready_rate",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "ordinary_instantiation_ready_count",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "strict_instantiation_ready_rate",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "strict_instantiation_ready_count",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "exact5_mean",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "exact20_mean",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "key_overlap_mean",
+      "nullable": true
+    }
+  ],
+  "patch_case_metrics": [
+    {
+      "arrow_type": "large_string",
+      "name": "example_uid",
+      "nullable": true
+    },
+    {
+      "arrow_type": "large_string",
+      "name": "changed_status",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "old_coverage",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "new_coverage",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "old_type_match",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "new_type_match",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "old_strict_ready",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "new_strict_ready",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "old_ordinary_ready",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "new_ordinary_ready",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "old_exact5",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "new_exact5",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "old_exact20",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "new_exact20",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "exact5_delta",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "exact20_delta",
+      "nullable": true
+    }
+  ],
+  "patch_transitions": [
+    {
+      "arrow_type": "large_string",
+      "name": "metric",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "both",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "prepatch_only",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "patched_only",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "neither",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "mcnemar_p",
+      "nullable": true
+    },
+    {
+      "arrow_type": "large_string",
+      "name": "mcnemar_note",
+      "nullable": true
+    }
+  ],
+  "per_query_metrics": [
+    {
+      "arrow_type": "large_string",
+      "name": "example_uid",
+      "nullable": true
+    },
+    {
+      "arrow_type": "large_string",
+      "name": "variant",
+      "nullable": true
+    },
+    {
+      "arrow_type": "large_string",
+      "name": "method",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "schema_hit",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "ordinary_ready",
+      "nullable": true
+    },
+    {
+      "arrow_type": "bool",
+      "name": "strict_ready",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "n_expected_scalar",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "n_filled",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "param_coverage",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "type_match",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "exact5",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "exact20",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "key_overlap",
+      "nullable": true
+    }
+  ],
+  "strict_readiness_summary": [
+    {
+      "arrow_type": "large_string",
+      "name": "metric",
+      "nullable": true
+    },
+    {
+      "arrow_type": "int64",
+      "name": "count",
+      "nullable": true
+    },
+    {
+      "arrow_type": "double",
+      "name": "rate",
+      "nullable": true
+    }
+  ]
+}

metadata/security_scan_report.json

@@ -0,0 +1,12 @@
+{
+  "explained_documentation_findings_count": 0,
+  "explained_findings_summary": "None",
+  "native_ids_included": false,
+  "parquet_string_field_scan": "PASS",
+  "private_mapping_in_public_staging": false,
+  "raw_text_included": false,
+  "scan_scope": "All public staging text files except this report, plus every Parquet string/schema field.",
+  "status": "PASS",
+  "unexplained_findings": [],
+  "unexplained_findings_count": 0
+}

metadata/statistics.json

@@ -0,0 +1,460 @@
+{
+  "method_summary": {
+    "columns": 13,
+    "null_counts": {
+      "coverage_mean": 0,
+      "exact20_mean": 0,
+      "exact5_mean": 0,
+      "key_overlap_mean": 0,
+      "method": 0,
+      "n": 0,
+      "ordinary_instantiation_ready_count": 0,
+      "ordinary_instantiation_ready_rate": 0,
+      "schema_correct_count": 0,
+      "schema_r1": 0,
+      "strict_instantiation_ready_count": 0,
+      "strict_instantiation_ready_rate": 0,
+      "type_match_mean": 0
+    },
+    "numeric_summary": {
+      "coverage_mean": {
+        "dtype": "float64",
+        "max": 0.888565596873754,
+        "min": 0.8794302206688913,
+        "unique_values_if_small": [
+          0.8794302206688913,
+          0.888565596873754
+        ]
+      },
+      "exact20_mean": {
+        "dtype": "float64",
+        "max": 0.2613912814943743,
+        "min": 0.2448882541666046,
+        "unique_values_if_small": [
+          0.2448882541666046,
+          0.2613912814943743
+        ]
+      },
+      "exact5_mean": {
+        "dtype": "float64",
+        "max": 0.2355361419794409,
+        "min": 0.2194626679162761,
+        "unique_values_if_small": [
+          0.2194626679162761,
+          0.2355361419794409
+        ]
+      },
+      "key_overlap_mean": {
+        "dtype": "float64",
+        "max": 0.9217810386994676,
+        "min": 0.9217810386994676,
+        "unique_values_if_small": [
+          0.9217810386994676
+        ]
+      },
+      "n": {
+        "dtype": "int64",
+        "max": 331.0,
+        "min": 331.0,
+        "unique_values_if_small": [
+          331
+        ]
+      },
+      "ordinary_instantiation_ready_count": {
+        "dtype": "int64",
+        "max": 265.0,
+        "min": 257.0,
+        "unique_values_if_small": [
+          257,
+          265
+        ]
+      },
+      "ordinary_instantiation_ready_rate": {
+        "dtype": "float64",
+        "max": 0.8006042296072508,
+        "min": 0.7764350453172205,
+        "unique_values_if_small": [
+          0.7764350453172205,
+          0.8006042296072508
+        ]
+      },
+      "schema_correct_count": {
+        "dtype": "int64",
+        "max": 301.0,
+        "min": 301.0,
+        "unique_values_if_small": [
+          301
+        ]
+      },
+      "schema_r1": {
+        "dtype": "float64",
+        "max": 0.9093655589123868,
+        "min": 0.9093655589123868,
+        "unique_values_if_small": [
+          0.9093655589123868
+        ]
+      },
+      "strict_instantiation_ready_count": {
+        "dtype": "int64",
+        "max": 255.0,
+        "min": 247.0,
+        "unique_values_if_small": [
+          247,
+          255
+        ]
+      },
+      "strict_instantiation_ready_rate": {
+        "dtype": "float64",
+        "max": 0.770392749244713,
+        "min": 0.7462235649546828,
+        "unique_values_if_small": [
+          0.7462235649546828,
+          0.770392749244713
+        ]
+      },
+      "type_match_mean": {
+        "dtype": "float64",
+        "max": 0.866549177576368,
+        "min": 0.8515453411979091,
+        "unique_values_if_small": [
+          0.8515453411979091,
+          0.866549177576368
+        ]
+      }
+    },
+    "rows": 2
+  },
+  "patch_case_metrics": {
+    "columns": 16,
+    "null_counts": {
+      "changed_status": 0,
+      "exact20_delta": 0,
+      "exact5_delta": 0,
+      "example_uid": 0,
+      "new_coverage": 0,
+      "new_exact20": 0,
+      "new_exact5": 0,
+      "new_ordinary_ready": 0,
+      "new_strict_ready": 0,
+      "new_type_match": 0,
+      "old_coverage": 0,
+      "old_exact20": 0,
+      "old_exact5": 0,
+      "old_ordinary_ready": 0,
+      "old_strict_ready": 0,
+      "old_type_match": 0
+    },
+    "numeric_summary": {
+      "exact20_delta": {
+        "dtype": "float64",
+        "max": 0.5,
+        "min": -0.1999999999999999,
+        "unique_values_if_small": null
+      },
+      "exact5_delta": {
+        "dtype": "float64",
+        "max": 0.5,
+        "min": -0.1999999999999999,
+        "unique_values_if_small": null
+      },
+      "new_coverage": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.8,
+        "unique_values_if_small": [
+          1.0,
+          0.8
+        ]
+      },
+      "new_exact20": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "new_exact5": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "new_ordinary_ready": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "new_strict_ready": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "new_type_match": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.6,
+        "unique_values_if_small": [
+          1.0,
+          0.6,
+          0.75,
+          0.8571428571428571,
+          0.875,
+          0.7142857142857143,
+          0.8
+        ]
+      },
+      "old_coverage": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.6,
+        "unique_values_if_small": [
+          0.8333333333333334,
+          0.8,
+          1.0,
+          0.8571428571428571,
+          0.75,
+          0.6,
+          0.875,
+          0.9
+        ]
+      },
+      "old_exact20": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "old_exact5": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "old_ordinary_ready": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "old_strict_ready": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "old_type_match": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.5,
+        "unique_values_if_small": null
+      }
+    },
+    "rows": 36
+  },
+  "patch_transitions": {
+    "columns": 7,
+    "null_counts": {
+      "both": 0,
+      "mcnemar_note": 0,
+      "mcnemar_p": 3,
+      "metric": 0,
+      "neither": 0,
+      "patched_only": 0,
+      "prepatch_only": 0
+    },
+    "numeric_summary": {
+      "both": {
+        "dtype": "int64",
+        "max": 301.0,
+        "min": 8.0,
+        "unique_values_if_small": [
+          247,
+          257,
+          301,
+          8
+        ]
+      },
+      "mcnemar_p": {
+        "dtype": "float64",
+        "max": 0.0078125,
+        "min": 0.0078125,
+        "unique_values_if_small": [
+          0.0078125
+        ]
+      },
+      "neither": {
+        "dtype": "int64",
+        "max": 323.0,
+        "min": 30.0,
+        "unique_values_if_small": [
+          76,
+          66,
+          30,
+          323
+        ]
+      },
+      "patched_only": {
+        "dtype": "int64",
+        "max": 8.0,
+        "min": 0.0,
+        "unique_values_if_small": [
+          8,
+          0
+        ]
+      },
+      "prepatch_only": {
+        "dtype": "int64",
+        "max": 0.0,
+        "min": 0.0,
+        "unique_values_if_small": [
+          0
+        ]
+      }
+    },
+    "rows": 5
+  },
+  "per_query_metrics": {
+    "columns": 13,
+    "null_counts": {
+      "exact20": 40,
+      "exact5": 40,
+      "example_uid": 0,
+      "key_overlap": 0,
+      "method": 0,
+      "n_expected_scalar": 0,
+      "n_filled": 0,
+      "ordinary_ready": 0,
+      "param_coverage": 0,
+      "schema_hit": 0,
+      "strict_ready": 0,
+      "type_match": 0,
+      "variant": 0
+    },
+    "numeric_summary": {
+      "exact20": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "exact5": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "key_overlap": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": [
+          1.0,
+          0.0,
+          0.4285714285714285,
+          0.7142857142857143,
+          0.8,
+          0.3333333333333333,
+          0.25,
+          0.5,
+          0.6666666666666666
+        ]
+      },
+      "n_expected_scalar": {
+        "dtype": "int64",
+        "max": 13.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "n_filled": {
+        "dtype": "int64",
+        "max": 12.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "ordinary_ready": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "param_coverage": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      },
+      "schema_hit": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "strict_ready": {
+        "dtype": "bool",
+        "max": true,
+        "min": false,
+        "unique_values_if_small": [
+          true,
+          false
+        ]
+      },
+      "type_match": {
+        "dtype": "float64",
+        "max": 1.0,
+        "min": 0.0,
+        "unique_values_if_small": null
+      }
+    },
+    "rows": 331
+  },
+  "strict_readiness_summary": {
+    "columns": 3,
+    "null_counts": {
+      "count": 0,
+      "metric": 0,
+      "rate": 0
+    },
+    "numeric_summary": {
+      "count": {
+        "dtype": "int64",
+        "max": 255.0,
+        "min": 8.0,
+        "unique_values_if_small": [
+          247,
+          255,
+          8
+        ]
+      },
+      "rate": {
+        "dtype": "float64",
+        "max": 0.770392749244713,
+        "min": 0.0241691842900302,
+        "unique_values_if_small": [
+          0.7462235649546828,
+          0.770392749244713,
+          0.0241691842900302
+        ]
+      }
+    },
+    "rows": 3
+  }
+}

metadata/upstream_attribution.md

@@ -0,0 +1,9 @@
+# Upstream Attribution
+
+This release uses NLP4LP as the upstream benchmark for evaluation. NLP4LP is a third-party benchmark/resource maintained by the Udell Lab and hosted at https://huggingface.co/datasets/udell-lab/NLP4LP. The observed Hugging Face license tag for NLP4LP during this build was cc-by-nc-sa-4.0, and access was gated.
+
+The NLP4LP/OptiMUS context should be cited as:
+
+Ahmaditeshnizi, Ali, Wenzhi Gao, and Madeleine Udell. OptiMUS: Scalable Optimization Modeling with (MI)LP Solvers and Large Language Models. Proceedings of the 41st International Conference on Machine Learning, PMLR 235:577-596, 2024. https://proceedings.mlr.press/v235/ahmaditeshnizi24a.html
+
+This dataset does not include or relicense raw NLP4LP problem text, gold formulations, gold structures, or source identifiers. Users needing the source benchmark should obtain NLP4LP directly from its canonical source under its own terms.