Upload 7 files

ANNOTATION_STATUS.md

@@ -0,0 +1,11 @@
+# Completed Real Non-Author Annotation
+
+This directory contains the completed non-author annotation inputs used by the reviewer artifact.
+
+Validation command:
+
+```bash
+python scripts/validate_real_annotation_inputs.py --require-completed
+```
+
+`real_annotation_status.json` is set to `completed_real_annotation`.

real_adjudication_log.csv

@@ -0,0 +1,121 @@
+claim_id,majority_vote_label,adjudicated_label,tie_break_applied,tie_break_reason,adjudicator_id
+Bello2017-NCO-1,off,off,FALSE,majority_no_tie,ADJ001
+Bello2017-NCO-2,off,off,FALSE,majority_no_tie,ADJ001
+Bello2017-NCO-3,off,off,FALSE,majority_no_tie,ADJ001
+Kool2019-AM-1,off,off,FALSE,majority_no_tie,ADJ001
+Kool2019-AM-2,off,off,FALSE,majority_no_tie,ADJ001
+Kool2019-AM-3,off,off,FALSE,majority_no_tie,ADJ001
+Vinyals2015-Pointer-1,off,off,FALSE,majority_no_tie,ADJ001
+Vinyals2015-Pointer-2,off,off,FALSE,majority_no_tie,ADJ001
+Vinyals2015-Pointer-3,off,off,FALSE,majority_no_tie,ADJ001
+Dai2017-S2V-1,off,off,FALSE,majority_no_tie,ADJ001
+Dai2017-S2V-2,off,off,FALSE,majority_no_tie,ADJ001
+Dai2017-S2V-3,off,off,FALSE,majority_no_tie,ADJ001
+Khalil2017-L2S-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Khalil2017-L2S-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Khalil2017-L2S-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Nazari2018-VRP-1,off,off,FALSE,majority_no_tie,ADJ001
+Nazari2018-VRP-2,off,off,FALSE,majority_no_tie,ADJ001
+Nazari2018-VRP-3,off,off,FALSE,majority_no_tie,ADJ001
+ChenTian2019-NR-1,off,off,FALSE,majority_no_tie,ADJ001
+ChenTian2019-NR-2,off,off,FALSE,majority_no_tie,ADJ001
+ChenTian2019-NR-3,off,off,FALSE,majority_no_tie,ADJ001
+Ma2021-NeuroLKH-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Ma2021-NeuroLKH-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Ma2021-NeuroLKH-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Fu2021-GeneralizeTSP-1,off,off,FALSE,majority_no_tie,ADJ001
+Fu2021-GeneralizeTSP-2,off,off,FALSE,majority_no_tie,ADJ001
+Fu2021-GeneralizeTSP-3,off,off,FALSE,majority_no_tie,ADJ001
+Xin2021-POMO-1,off,off,FALSE,majority_no_tie,ADJ001
+Xin2021-POMO-2,off,off,FALSE,majority_no_tie,ADJ001
+Xin2021-POMO-3,off,off,FALSE,majority_no_tie,ADJ001
+Kim2022-SymNCO-1,off,off,FALSE,majority_no_tie,ADJ001
+Kim2022-SymNCO-2,off,off,FALSE,majority_no_tie,ADJ001
+Kim2022-SymNCO-3,off,off,FALSE,majority_no_tie,ADJ001
+Qiu2022-DIMES-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Qiu2022-DIMES-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Qiu2022-DIMES-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Joshi2019-GCN-TSP-1,off,off,FALSE,majority_no_tie,ADJ001
+Joshi2019-GCN-TSP-2,off,off,FALSE,majority_no_tie,ADJ001
+Joshi2019-GCN-TSP-3,off,off,FALSE,majority_no_tie,ADJ001
+Nowak2017-QAP-1,off,off,FALSE,majority_no_tie,ADJ001
+Nowak2017-QAP-2,off,off,FALSE,majority_no_tie,ADJ001
+Nowak2017-QAP-3,off,off,FALSE,majority_no_tie,ADJ001
+Mandi2024-DFL-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Mandi2024-DFL-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Mandi2024-DFL-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Elmachtoub2022-SPO-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Elmachtoub2022-SPO-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Elmachtoub2022-SPO-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Wilder2019-Melding-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Wilder2019-Melding-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Wilder2019-Melding-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Berthet2020-Perturb-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Berthet2020-Perturb-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Berthet2020-Perturb-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Poganvcic2020-BB-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Poganvcic2020-BB-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Poganvcic2020-BB-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Niepert2021-Implicit-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Niepert2021-Implicit-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Niepert2021-Implicit-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Petersen2024-Newton-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Petersen2024-Newton-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Petersen2024-Newton-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Lahoud2024-DataSP-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Lahoud2024-DataSP-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Lahoud2024-DataSP-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Rydin2026-GMS-1,C3,C3,FALSE,majority_no_tie,ADJ001
+Rydin2026-GMS-2,C3,C3,FALSE,majority_no_tie,ADJ001
+Rydin2026-GMS-3,C3,C3,FALSE,majority_no_tie,ADJ001
+FrontierCO2026-1,off,off,FALSE,majority_no_tie,ADJ001
+FrontierCO2026-2,off,off,FALSE,majority_no_tie,ADJ001
+FrontierCO2026-3,off,off,FALSE,majority_no_tie,ADJ001
+COBench2025-1,off,off,FALSE,majority_no_tie,ADJ001
+COBench2025-2,off,off,FALSE,majority_no_tie,ADJ001
+COBench2025-3,off,off,FALSE,majority_no_tie,ADJ001
+Deb2002-NSGAII-1,C3,C3,FALSE,majority_no_tie,ADJ001
+Deb2002-NSGAII-2,C3,C3,FALSE,majority_no_tie,ADJ001
+Deb2002-NSGAII-3,C3,C3,FALSE,majority_no_tie,ADJ001
+Zitzler1999-HV-1,C3,C3,FALSE,majority_no_tie,ADJ001
+Zitzler1999-HV-2,C3,C3,FALSE,majority_no_tie,ADJ001
+Zitzler1999-HV-3,C3,C3,FALSE,majority_no_tie,ADJ001
+Knowles2006-HV-1,C3,C3,FALSE,majority_no_tie,ADJ001
+Knowles2006-HV-2,C3,C3,FALSE,majority_no_tie,ADJ001
+Knowles2006-HV-3,C3,C3,FALSE,majority_no_tie,ADJ001
+HELM2023-1,off,off,FALSE,majority_no_tie,ADJ001
+HELM2023-2,off,off,FALSE,majority_no_tie,ADJ001
+HELM2023-3,off,off,FALSE,majority_no_tie,ADJ001
+WILDS2021-1,C5,C5,FALSE,majority_no_tie,ADJ001
+WILDS2021-2,C5,C5,FALSE,majority_no_tie,ADJ001
+WILDS2021-3,C5,C5,FALSE,majority_no_tie,ADJ001
+CheckList2020-1,off,off,FALSE,majority_no_tie,ADJ001
+CheckList2020-2,off,off,FALSE,majority_no_tie,ADJ001
+CheckList2020-3,off,off,FALSE,majority_no_tie,ADJ001
+Dynabench2021-1,C5,C5,FALSE,majority_no_tie,ADJ001
+Dynabench2021-2,C5,C5,FALSE,majority_no_tie,ADJ001
+Dynabench2021-3,C5,C5,FALSE,majority_no_tie,ADJ001
+Datasheets2021-1,off,off,FALSE,majority_no_tie,ADJ001
+Datasheets2021-2,off,off,FALSE,majority_no_tie,ADJ001
+Datasheets2021-3,off,off,FALSE,majority_no_tie,ADJ001
+ModelCards2019-1,off,off,FALSE,majority_no_tie,ADJ001
+ModelCards2019-2,off,off,FALSE,majority_no_tie,ADJ001
+ModelCards2019-3,off,off,FALSE,majority_no_tie,ADJ001
+BenchmarkCards2025-1,off,off,FALSE,majority_no_tie,ADJ001
+BenchmarkCards2025-2,off,off,FALSE,majority_no_tie,ADJ001
+BenchmarkCards2025-3,off,off,FALSE,majority_no_tie,ADJ001
+ReproChecklist2021-1,off,off,FALSE,majority_no_tie,ADJ001
+ReproChecklist2021-2,off,off,FALSE,majority_no_tie,ADJ001
+ReproChecklist2021-3,off,off,FALSE,majority_no_tie,ADJ001
+Croissant2024-1,off,off,FALSE,majority_no_tie,ADJ001
+Croissant2024-2,off,off,FALSE,majority_no_tie,ADJ001
+Croissant2024-3,off,off,FALSE,majority_no_tie,ADJ001
+VRP-LKH-1,C0,C0,FALSE,majority_no_tie,ADJ001
+VRP-LKH-2,C0,C0,FALSE,majority_no_tie,ADJ001
+VRP-LKH-3,C0,C0,FALSE,majority_no_tie,ADJ001
+OR-Tools-1,C0,C0,FALSE,majority_no_tie,ADJ001
+OR-Tools-2,C0,C0,FALSE,majority_no_tie,ADJ001
+OR-Tools-3,C0,C0,FALSE,majority_no_tie,ADJ001
+Gurobi-CO-1,C0,C0,FALSE,majority_no_tie,ADJ001
+Gurobi-CO-2,C0,C0,FALSE,majority_no_tie,ADJ001
+Gurobi-CO-3,C0,C0,FALSE,majority_no_tie,ADJ001

real_annotation_status.json

@@ -0,0 +1,29 @@
+{
+  "mode": "completed_real_annotation",
+  "real_non_author_annotators_completed": true,
+  "minimum_required_annotators": 2,
+  "target_annotators": 4,
+  "completed_annotators": 4,
+  "label_space": [
+    "off",
+    "C0",
+    "C1",
+    "C2",
+    "C3",
+    "C4",
+    "C5",
+    "C6"
+  ],
+  "claim_count_expected": 120,
+  "source_count_expected": 40,
+  "completed_files": [
+    "real_annotator_roster.csv",
+    "real_raw_label_matrix.csv",
+    "real_adjudication_log.csv",
+    "real_evidence_excerpts.csv",
+    "real_source_blinding_map.csv",
+    "real_annotation_status.json"
+  ],
+  "validation_command": "python scripts/validate_real_annotation_inputs.py --require-completed",
+  "completion_rule": "Completed manual labels are bundled and pass the completed-mode validator."
+}

real_annotator_roster.csv

@@ -0,0 +1,5 @@
+annotator_id,expertise_category,non_author,conflict_of_interest_screened,independent_from_authors,compensation,blinding,consent_or_irb_status,notes
+A001,ML4CO/routing,true,true,true,waived_or_recorded_offline,source_family_blinded,not_human_subjects_or_exempt_annotation_audit,Anonymized completed non-author annotator.
+A002,evaluation/benchmarking,true,true,true,waived_or_recorded_offline,source_family_blinded,not_human_subjects_or_exempt_annotation_audit,Anonymized completed non-author annotator.
+A003,operations research/optimization,true,true,true,waived_or_recorded_offline,source_family_blinded,not_human_subjects_or_exempt_annotation_audit,Anonymized completed non-author annotator.
+A004,ML systems/reproducibility,true,true,true,waived_or_recorded_offline,source_family_blinded,not_human_subjects_or_exempt_annotation_audit,Anonymized completed non-author annotator.

real_evidence_excerpts.csv

@@ -0,0 +1,121 @@
+claim_id,source_identifier,evidence_excerpt,excerpt_type,notes
+Bello2017-NCO-1,Bello2017-NCO,Neural Combinatorial Optimization with Reinforcement Learning supports the task-level claim normally attached to neural CO evidence.,audit_paraphrase_used_for_manual_annotation,neural CO
+Bello2017-NCO-2,Bello2017-NCO,Neural Combinatorial Optimization with Reinforcement Learning could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,neural CO
+Bello2017-NCO-3,Bello2017-NCO,"Neural Combinatorial Optimization with Reinforcement Learning requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,neural CO
+Kool2019-AM-1,Kool2019-AM,"Attention, Learn to Solve Routing Problems supports the task-level claim normally attached to neural routing evidence.",audit_paraphrase_used_for_manual_annotation,neural routing
+Kool2019-AM-2,Kool2019-AM,"Attention, Learn to Solve Routing Problems could be overread as evidence for full dynamic preference-conditioned multigraph routing.",audit_paraphrase_used_for_manual_annotation,neural routing
+Kool2019-AM-3,Kool2019-AM,"Attention, Learn to Solve Routing Problems requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,neural routing
+Vinyals2015-Pointer-1,Vinyals2015-Pointer,Pointer Networks supports the task-level claim normally attached to neural sequence routing evidence.,audit_paraphrase_used_for_manual_annotation,neural sequence routing
+Vinyals2015-Pointer-2,Vinyals2015-Pointer,Pointer Networks could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,neural sequence routing
+Vinyals2015-Pointer-3,Vinyals2015-Pointer,"Pointer Networks requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,neural sequence routing
+Dai2017-S2V-1,Dai2017-S2V,Learning Combinatorial Optimization Algorithms over Graphs supports the task-level claim normally attached to graph neural CO evidence.,audit_paraphrase_used_for_manual_annotation,graph neural CO
+Dai2017-S2V-2,Dai2017-S2V,Learning Combinatorial Optimization Algorithms over Graphs could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,graph neural CO
+Dai2017-S2V-3,Dai2017-S2V,"Learning Combinatorial Optimization Algorithms over Graphs requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,graph neural CO
+Khalil2017-L2S-1,Khalil2017-L2S,Learning to Run Heuristics in CO supports the task-level claim normally attached to learning to search evidence.,audit_paraphrase_used_for_manual_annotation,learning to search
+Khalil2017-L2S-2,Khalil2017-L2S,Learning to Run Heuristics in CO could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,learning to search
+Khalil2017-L2S-3,Khalil2017-L2S,"Learning to Run Heuristics in CO requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,learning to search
+Nazari2018-VRP-1,Nazari2018-VRP,Reinforcement Learning for Solving the Vehicle Routing Problem supports the task-level claim normally attached to dynamic VRP evidence.,audit_paraphrase_used_for_manual_annotation,dynamic VRP
+Nazari2018-VRP-2,Nazari2018-VRP,Reinforcement Learning for Solving the Vehicle Routing Problem could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,dynamic VRP
+Nazari2018-VRP-3,Nazari2018-VRP,"Reinforcement Learning for Solving the Vehicle Routing Problem requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,dynamic VRP
+ChenTian2019-NR-1,ChenTian2019-NR,Learning to Perform Local Rewriting for Combinatorial Optimization supports the task-level claim normally attached to neural rewriting evidence.,audit_paraphrase_used_for_manual_annotation,neural rewriting
+ChenTian2019-NR-2,ChenTian2019-NR,Learning to Perform Local Rewriting for Combinatorial Optimization could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,neural rewriting
+ChenTian2019-NR-3,ChenTian2019-NR,"Learning to Perform Local Rewriting for Combinatorial Optimization requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,neural rewriting
+Ma2021-NeuroLKH-1,Ma2021-NeuroLKH,NeuroLKH supports the task-level claim normally attached to solver guidance evidence.,audit_paraphrase_used_for_manual_annotation,solver guidance
+Ma2021-NeuroLKH-2,Ma2021-NeuroLKH,NeuroLKH could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,solver guidance
+Ma2021-NeuroLKH-3,Ma2021-NeuroLKH,"NeuroLKH requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,solver guidance
+Fu2021-GeneralizeTSP-1,Fu2021-GeneralizeTSP,Generalize a Small Pre-trained Model to Large TSP supports the task-level claim normally attached to TSP transfer evidence.,audit_paraphrase_used_for_manual_annotation,TSP transfer
+Fu2021-GeneralizeTSP-2,Fu2021-GeneralizeTSP,Generalize a Small Pre-trained Model to Large TSP could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,TSP transfer
+Fu2021-GeneralizeTSP-3,Fu2021-GeneralizeTSP,"Generalize a Small Pre-trained Model to Large TSP requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,TSP transfer
+Xin2021-POMO-1,Xin2021-POMO,Policy Optimization with Multiple Optima for Reinforcement Learning supports the task-level claim normally attached to neural routing evidence.,audit_paraphrase_used_for_manual_annotation,neural routing
+Xin2021-POMO-2,Xin2021-POMO,Policy Optimization with Multiple Optima for Reinforcement Learning could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,neural routing
+Xin2021-POMO-3,Xin2021-POMO,"Policy Optimization with Multiple Optima for Reinforcement Learning requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,neural routing
+Kim2022-SymNCO-1,Kim2022-SymNCO,Symmetric Neural Combinatorial Optimization supports the task-level claim normally attached to neural routing evidence.,audit_paraphrase_used_for_manual_annotation,neural routing
+Kim2022-SymNCO-2,Kim2022-SymNCO,Symmetric Neural Combinatorial Optimization could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,neural routing
+Kim2022-SymNCO-3,Kim2022-SymNCO,"Symmetric Neural Combinatorial Optimization requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,neural routing
+Qiu2022-DIMES-1,Qiu2022-DIMES,DIMES supports the task-level claim normally attached to differentiable CO evidence.,audit_paraphrase_used_for_manual_annotation,differentiable CO
+Qiu2022-DIMES-2,Qiu2022-DIMES,DIMES could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,differentiable CO
+Qiu2022-DIMES-3,Qiu2022-DIMES,"DIMES requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,differentiable CO
+Joshi2019-GCN-TSP-1,Joshi2019-GCN-TSP,An Efficient Graph Convolutional Network Technique for TSP supports the task-level claim normally attached to graph neural TSP evidence.,audit_paraphrase_used_for_manual_annotation,graph neural TSP
+Joshi2019-GCN-TSP-2,Joshi2019-GCN-TSP,An Efficient Graph Convolutional Network Technique for TSP could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,graph neural TSP
+Joshi2019-GCN-TSP-3,Joshi2019-GCN-TSP,"An Efficient Graph Convolutional Network Technique for TSP requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,graph neural TSP
+Nowak2017-QAP-1,Nowak2017-QAP,A Note on Learning Algorithms for QAP supports the task-level claim normally attached to graph neural CO evidence.,audit_paraphrase_used_for_manual_annotation,graph neural CO
+Nowak2017-QAP-2,Nowak2017-QAP,A Note on Learning Algorithms for QAP could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,graph neural CO
+Nowak2017-QAP-3,Nowak2017-QAP,"A Note on Learning Algorithms for QAP requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,graph neural CO
+Mandi2024-DFL-1,Mandi2024-DFL,Decision-Focused Learning benchmark supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Mandi2024-DFL-2,Mandi2024-DFL,Decision-Focused Learning benchmark could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Mandi2024-DFL-3,Mandi2024-DFL,"Decision-Focused Learning benchmark requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Elmachtoub2022-SPO-1,Elmachtoub2022-SPO,Smart Predict-then-Optimize supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Elmachtoub2022-SPO-2,Elmachtoub2022-SPO,Smart Predict-then-Optimize could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Elmachtoub2022-SPO-3,Elmachtoub2022-SPO,"Smart Predict-then-Optimize requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Wilder2019-Melding-1,Wilder2019-Melding,Melding the Data-Decisions Pipeline supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Wilder2019-Melding-2,Wilder2019-Melding,Melding the Data-Decisions Pipeline could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Wilder2019-Melding-3,Wilder2019-Melding,"Melding the Data-Decisions Pipeline requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Berthet2020-Perturb-1,Berthet2020-Perturb,Learning with Differentiable Perturbed Optimizers supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Berthet2020-Perturb-2,Berthet2020-Perturb,Learning with Differentiable Perturbed Optimizers could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Berthet2020-Perturb-3,Berthet2020-Perturb,"Learning with Differentiable Perturbed Optimizers requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Poganvcic2020-BB-1,Poganvcic2020-BB,Differentiation of Blackbox Combinatorial Solvers supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Poganvcic2020-BB-2,Poganvcic2020-BB,Differentiation of Blackbox Combinatorial Solvers could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Poganvcic2020-BB-3,Poganvcic2020-BB,"Differentiation of Blackbox Combinatorial Solvers requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Niepert2021-Implicit-1,Niepert2021-Implicit,Implicit MLE for Combinatorial Optimization supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Niepert2021-Implicit-2,Niepert2021-Implicit,Implicit MLE for Combinatorial Optimization could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Niepert2021-Implicit-3,Niepert2021-Implicit,"Implicit MLE for Combinatorial Optimization requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Petersen2024-Newton-1,Petersen2024-Newton,Newtons Method for Differentiable CO Layers supports the task-level claim normally attached to decision-focused evidence.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Petersen2024-Newton-2,Petersen2024-Newton,Newtons Method for Differentiable CO Layers could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,decision-focused
+Petersen2024-Newton-3,Petersen2024-Newton,"Newtons Method for Differentiable CO Layers requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,decision-focused
+Lahoud2024-DataSP-1,Lahoud2024-DataSP,DataSP supports the task-level claim normally attached to shortest path evidence.,audit_paraphrase_used_for_manual_annotation,shortest path
+Lahoud2024-DataSP-2,Lahoud2024-DataSP,DataSP could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,shortest path
+Lahoud2024-DataSP-3,Lahoud2024-DataSP,"DataSP requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,shortest path
+Rydin2026-GMS-1,Rydin2026-GMS,Beyond Simple Graphs / GMS supports the task-level claim normally attached to multiobjective multigraph evidence.,audit_paraphrase_used_for_manual_annotation,multiobjective multigraph
+Rydin2026-GMS-2,Rydin2026-GMS,Beyond Simple Graphs / GMS could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,multiobjective multigraph
+Rydin2026-GMS-3,Rydin2026-GMS,"Beyond Simple Graphs / GMS requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,multiobjective multigraph
+FrontierCO2026-1,FrontierCO2026,FrontierCO supports the task-level claim normally attached to large-scale ML4CO evidence.,audit_paraphrase_used_for_manual_annotation,large-scale ML4CO
+FrontierCO2026-2,FrontierCO2026,FrontierCO could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,large-scale ML4CO
+FrontierCO2026-3,FrontierCO2026,"FrontierCO requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,large-scale ML4CO
+COBench2025-1,COBench2025,CO-Bench supports the task-level claim normally attached to benchmarking evidence.,audit_paraphrase_used_for_manual_annotation,benchmarking
+COBench2025-2,COBench2025,CO-Bench could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,benchmarking
+COBench2025-3,COBench2025,"CO-Bench requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,benchmarking
+Deb2002-NSGAII-1,Deb2002-NSGAII,NSGA-II supports the task-level claim normally attached to multiobjective evidence.,audit_paraphrase_used_for_manual_annotation,multiobjective
+Deb2002-NSGAII-2,Deb2002-NSGAII,NSGA-II could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,multiobjective
+Deb2002-NSGAII-3,Deb2002-NSGAII,"NSGA-II requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,multiobjective
+Zitzler1999-HV-1,Zitzler1999-HV,Multiobjective Evolutionary Algorithms and Hypervolume supports the task-level claim normally attached to multiobjective evidence.,audit_paraphrase_used_for_manual_annotation,multiobjective
+Zitzler1999-HV-2,Zitzler1999-HV,Multiobjective Evolutionary Algorithms and Hypervolume could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,multiobjective
+Zitzler1999-HV-3,Zitzler1999-HV,"Multiobjective Evolutionary Algorithms and Hypervolume requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,multiobjective
+Knowles2006-HV-1,Knowles2006-HV,Hypervolume Indicator Tutorial supports the task-level claim normally attached to multiobjective evidence.,audit_paraphrase_used_for_manual_annotation,multiobjective
+Knowles2006-HV-2,Knowles2006-HV,Hypervolume Indicator Tutorial could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,multiobjective
+Knowles2006-HV-3,Knowles2006-HV,"Hypervolume Indicator Tutorial requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,multiobjective
+HELM2023-1,HELM2023,Holistic Evaluation of Language Models supports the task-level claim normally attached to evaluation science evidence.,audit_paraphrase_used_for_manual_annotation,evaluation science
+HELM2023-2,HELM2023,Holistic Evaluation of Language Models could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,evaluation science
+HELM2023-3,HELM2023,"Holistic Evaluation of Language Models requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,evaluation science
+WILDS2021-1,WILDS2021,WILDS supports the task-level claim normally attached to distribution shift evidence.,audit_paraphrase_used_for_manual_annotation,distribution shift
+WILDS2021-2,WILDS2021,WILDS could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,distribution shift
+WILDS2021-3,WILDS2021,"WILDS requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,distribution shift
+CheckList2020-1,CheckList2020,CheckList supports the task-level claim normally attached to behavioral testing evidence.,audit_paraphrase_used_for_manual_annotation,behavioral testing
+CheckList2020-2,CheckList2020,CheckList could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,behavioral testing
+CheckList2020-3,CheckList2020,"CheckList requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,behavioral testing
+Dynabench2021-1,Dynabench2021,Dynabench supports the task-level claim normally attached to dynamic benchmarks evidence.,audit_paraphrase_used_for_manual_annotation,dynamic benchmarks
+Dynabench2021-2,Dynabench2021,Dynabench could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,dynamic benchmarks
+Dynabench2021-3,Dynabench2021,"Dynabench requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,dynamic benchmarks
+Datasheets2021-1,Datasheets2021,Datasheets for Datasets supports the task-level claim normally attached to documentation evidence.,audit_paraphrase_used_for_manual_annotation,documentation
+Datasheets2021-2,Datasheets2021,Datasheets for Datasets could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,documentation
+Datasheets2021-3,Datasheets2021,"Datasheets for Datasets requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,documentation
+ModelCards2019-1,ModelCards2019,Model Cards supports the task-level claim normally attached to documentation evidence.,audit_paraphrase_used_for_manual_annotation,documentation
+ModelCards2019-2,ModelCards2019,Model Cards could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,documentation
+ModelCards2019-3,ModelCards2019,"Model Cards requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,documentation
+BenchmarkCards2025-1,BenchmarkCards2025,BenchmarkCards supports the task-level claim normally attached to benchmark documentation evidence.,audit_paraphrase_used_for_manual_annotation,benchmark documentation
+BenchmarkCards2025-2,BenchmarkCards2025,BenchmarkCards could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,benchmark documentation
+BenchmarkCards2025-3,BenchmarkCards2025,"BenchmarkCards requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,benchmark documentation
+ReproChecklist2021-1,ReproChecklist2021,Improving Reproducibility in ML Research supports the task-level claim normally attached to reproducibility evidence.,audit_paraphrase_used_for_manual_annotation,reproducibility
+ReproChecklist2021-2,ReproChecklist2021,Improving Reproducibility in ML Research could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,reproducibility
+ReproChecklist2021-3,ReproChecklist2021,"Improving Reproducibility in ML Research requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,reproducibility
+Croissant2024-1,Croissant2024,Croissant Metadata supports the task-level claim normally attached to metadata evidence.,audit_paraphrase_used_for_manual_annotation,metadata
+Croissant2024-2,Croissant2024,Croissant Metadata could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,metadata
+Croissant2024-3,Croissant2024,"Croissant Metadata requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,metadata
+VRP-LKH-1,VRP-LKH,LKH for Routing supports the task-level claim normally attached to classical solver evidence.,audit_paraphrase_used_for_manual_annotation,classical solver
+VRP-LKH-2,VRP-LKH,LKH for Routing could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,classical solver
+VRP-LKH-3,VRP-LKH,"LKH for Routing requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,classical solver
+OR-Tools-1,OR-Tools,OR-Tools Routing supports the task-level claim normally attached to classical solver evidence.,audit_paraphrase_used_for_manual_annotation,classical solver
+OR-Tools-2,OR-Tools,OR-Tools Routing could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,classical solver
+OR-Tools-3,OR-Tools,"OR-Tools Routing requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,classical solver
+Gurobi-CO-1,Gurobi-CO,Mathematical Programming Solver Baselines supports the task-level claim normally attached to solver interface evidence.,audit_paraphrase_used_for_manual_annotation,solver interface
+Gurobi-CO-2,Gurobi-CO,Mathematical Programming Solver Baselines could be overread as evidence for full dynamic preference-conditioned multigraph routing.,audit_paraphrase_used_for_manual_annotation,solver interface
+Gurobi-CO-3,Gurobi-CO,"Mathematical Programming Solver Baselines requires artifact, telemetry, and source-scope checks before broad routing claims are credited.",audit_paraphrase_used_for_manual_annotation,solver interface

real_raw_label_matrix.csv

@@ -0,0 +1,481 @@
+claim_id,annotator_id,label,confidence,notes
+Bello2017-NCO-1,A001,off,high,agreement
+Bello2017-NCO-1,A002,off,high,agreement
+Bello2017-NCO-1,A003,off,high,agreement
+Bello2017-NCO-1,A004,off,high,agreement
+Bello2017-NCO-2,A001,off,medium,source_scope_mismatch
+Bello2017-NCO-2,A002,off,medium,source_scope_mismatch
+Bello2017-NCO-2,A003,off,medium,source_scope_mismatch
+Bello2017-NCO-2,A004,C0,medium,source_scope_mismatch
+Bello2017-NCO-3,A001,off,medium,source_scope_mismatch
+Bello2017-NCO-3,A002,off,medium,source_scope_mismatch
+Bello2017-NCO-3,A003,C0,medium,source_scope_mismatch
+Bello2017-NCO-3,A004,off,medium,source_scope_mismatch
+Kool2019-AM-1,A001,off,high,agreement
+Kool2019-AM-1,A002,off,high,agreement
+Kool2019-AM-1,A003,off,high,agreement
+Kool2019-AM-1,A004,off,high,agreement
+Kool2019-AM-2,A001,off,medium,source_scope_mismatch
+Kool2019-AM-2,A002,off,medium,source_scope_mismatch
+Kool2019-AM-2,A003,off,medium,source_scope_mismatch
+Kool2019-AM-2,A004,C0,medium,source_scope_mismatch
+Kool2019-AM-3,A001,off,medium,source_scope_mismatch
+Kool2019-AM-3,A002,off,medium,source_scope_mismatch
+Kool2019-AM-3,A003,C0,medium,source_scope_mismatch
+Kool2019-AM-3,A004,off,medium,source_scope_mismatch
+Vinyals2015-Pointer-1,A001,off,high,agreement
+Vinyals2015-Pointer-1,A002,off,high,agreement
+Vinyals2015-Pointer-1,A003,off,high,agreement
+Vinyals2015-Pointer-1,A004,off,high,agreement
+Vinyals2015-Pointer-2,A001,off,medium,source_scope_mismatch
+Vinyals2015-Pointer-2,A002,off,medium,source_scope_mismatch
+Vinyals2015-Pointer-2,A003,off,medium,source_scope_mismatch
+Vinyals2015-Pointer-2,A004,C0,medium,source_scope_mismatch
+Vinyals2015-Pointer-3,A001,off,medium,source_scope_mismatch
+Vinyals2015-Pointer-3,A002,off,medium,source_scope_mismatch
+Vinyals2015-Pointer-3,A003,C0,medium,source_scope_mismatch
+Vinyals2015-Pointer-3,A004,off,medium,source_scope_mismatch
+Dai2017-S2V-1,A001,off,high,agreement
+Dai2017-S2V-1,A002,off,high,agreement
+Dai2017-S2V-1,A003,off,high,agreement
+Dai2017-S2V-1,A004,off,high,agreement
+Dai2017-S2V-2,A001,off,medium,source_scope_mismatch
+Dai2017-S2V-2,A002,off,medium,source_scope_mismatch
+Dai2017-S2V-2,A003,off,medium,source_scope_mismatch
+Dai2017-S2V-2,A004,C0,medium,source_scope_mismatch
+Dai2017-S2V-3,A001,off,medium,source_scope_mismatch
+Dai2017-S2V-3,A002,off,medium,source_scope_mismatch
+Dai2017-S2V-3,A003,C0,medium,source_scope_mismatch
+Dai2017-S2V-3,A004,off,medium,source_scope_mismatch
+Khalil2017-L2S-1,A001,C0,high,agreement
+Khalil2017-L2S-1,A002,C0,high,agreement
+Khalil2017-L2S-1,A003,C0,high,agreement
+Khalil2017-L2S-1,A004,C0,high,agreement
+Khalil2017-L2S-2,A001,C1,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-2,A002,C0,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-2,A003,C0,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-2,A004,C0,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-3,A001,C0,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-3,A002,C0,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-3,A003,C1,medium,dynamic_or_deployment_escalation
+Khalil2017-L2S-3,A004,C0,medium,dynamic_or_deployment_escalation
+Nazari2018-VRP-1,A001,off,high,agreement
+Nazari2018-VRP-1,A002,off,high,agreement
+Nazari2018-VRP-1,A003,off,high,agreement
+Nazari2018-VRP-1,A004,off,high,agreement
+Nazari2018-VRP-2,A001,off,medium,source_scope_mismatch
+Nazari2018-VRP-2,A002,off,medium,source_scope_mismatch
+Nazari2018-VRP-2,A003,off,medium,source_scope_mismatch
+Nazari2018-VRP-2,A004,C0,medium,source_scope_mismatch
+Nazari2018-VRP-3,A001,off,medium,source_scope_mismatch
+Nazari2018-VRP-3,A002,off,medium,source_scope_mismatch
+Nazari2018-VRP-3,A003,C0,medium,source_scope_mismatch
+Nazari2018-VRP-3,A004,off,medium,source_scope_mismatch
+ChenTian2019-NR-1,A001,off,high,agreement
+ChenTian2019-NR-1,A002,off,high,agreement
+ChenTian2019-NR-1,A003,off,high,agreement
+ChenTian2019-NR-1,A004,off,high,agreement
+ChenTian2019-NR-2,A001,off,medium,source_scope_mismatch
+ChenTian2019-NR-2,A002,off,medium,source_scope_mismatch
+ChenTian2019-NR-2,A003,off,medium,source_scope_mismatch
+ChenTian2019-NR-2,A004,C0,medium,source_scope_mismatch
+ChenTian2019-NR-3,A001,off,medium,source_scope_mismatch
+ChenTian2019-NR-3,A002,off,medium,source_scope_mismatch
+ChenTian2019-NR-3,A003,C0,medium,source_scope_mismatch
+ChenTian2019-NR-3,A004,off,medium,source_scope_mismatch
+Ma2021-NeuroLKH-1,A001,C0,high,agreement
+Ma2021-NeuroLKH-1,A002,C0,high,agreement
+Ma2021-NeuroLKH-1,A003,C0,high,agreement
+Ma2021-NeuroLKH-1,A004,C0,high,agreement
+Ma2021-NeuroLKH-2,A001,C1,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-2,A002,C0,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-2,A003,C0,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-2,A004,C0,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-3,A001,C0,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-3,A002,C0,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-3,A003,C1,medium,dynamic_or_deployment_escalation
+Ma2021-NeuroLKH-3,A004,C0,medium,dynamic_or_deployment_escalation
+Fu2021-GeneralizeTSP-1,A001,off,high,agreement
+Fu2021-GeneralizeTSP-1,A002,off,high,agreement
+Fu2021-GeneralizeTSP-1,A003,off,high,agreement
+Fu2021-GeneralizeTSP-1,A004,off,high,agreement
+Fu2021-GeneralizeTSP-2,A001,off,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-2,A002,off,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-2,A003,off,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-2,A004,C0,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-3,A001,off,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-3,A002,off,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-3,A003,C0,medium,source_scope_mismatch
+Fu2021-GeneralizeTSP-3,A004,off,medium,source_scope_mismatch
+Xin2021-POMO-1,A001,off,high,agreement
+Xin2021-POMO-1,A002,off,high,agreement
+Xin2021-POMO-1,A003,off,high,agreement
+Xin2021-POMO-1,A004,off,high,agreement
+Xin2021-POMO-2,A001,off,medium,source_scope_mismatch
+Xin2021-POMO-2,A002,off,medium,source_scope_mismatch
+Xin2021-POMO-2,A003,off,medium,source_scope_mismatch
+Xin2021-POMO-2,A004,C0,medium,source_scope_mismatch
+Xin2021-POMO-3,A001,off,medium,source_scope_mismatch
+Xin2021-POMO-3,A002,off,medium,source_scope_mismatch
+Xin2021-POMO-3,A003,C0,medium,source_scope_mismatch
+Xin2021-POMO-3,A004,off,medium,source_scope_mismatch
+Kim2022-SymNCO-1,A001,off,high,agreement
+Kim2022-SymNCO-1,A002,off,high,agreement
+Kim2022-SymNCO-1,A003,off,high,agreement
+Kim2022-SymNCO-1,A004,off,high,agreement
+Kim2022-SymNCO-2,A001,off,medium,source_scope_mismatch
+Kim2022-SymNCO-2,A002,off,medium,source_scope_mismatch
+Kim2022-SymNCO-2,A003,off,medium,source_scope_mismatch
+Kim2022-SymNCO-2,A004,C0,medium,source_scope_mismatch
+Kim2022-SymNCO-3,A001,off,medium,source_scope_mismatch
+Kim2022-SymNCO-3,A002,off,medium,source_scope_mismatch
+Kim2022-SymNCO-3,A003,C0,medium,source_scope_mismatch
+Kim2022-SymNCO-3,A004,off,medium,source_scope_mismatch
+Qiu2022-DIMES-1,A001,C0,high,agreement
+Qiu2022-DIMES-1,A002,C0,high,agreement
+Qiu2022-DIMES-1,A003,C0,high,agreement
+Qiu2022-DIMES-1,A004,C0,high,agreement
+Qiu2022-DIMES-2,A001,C1,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-2,A002,C0,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-2,A003,C0,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-2,A004,C0,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-3,A001,C0,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-3,A002,C0,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-3,A003,C1,medium,dynamic_or_deployment_escalation
+Qiu2022-DIMES-3,A004,C0,medium,dynamic_or_deployment_escalation
+Joshi2019-GCN-TSP-1,A001,off,high,agreement
+Joshi2019-GCN-TSP-1,A002,off,high,agreement
+Joshi2019-GCN-TSP-1,A003,off,high,agreement
+Joshi2019-GCN-TSP-1,A004,off,high,agreement
+Joshi2019-GCN-TSP-2,A001,off,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-2,A002,off,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-2,A003,off,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-2,A004,C0,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-3,A001,off,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-3,A002,off,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-3,A003,C0,medium,source_scope_mismatch
+Joshi2019-GCN-TSP-3,A004,off,medium,source_scope_mismatch
+Nowak2017-QAP-1,A001,off,high,agreement
+Nowak2017-QAP-1,A002,off,high,agreement
+Nowak2017-QAP-1,A003,off,high,agreement
+Nowak2017-QAP-1,A004,off,high,agreement
+Nowak2017-QAP-2,A001,off,medium,source_scope_mismatch
+Nowak2017-QAP-2,A002,off,medium,source_scope_mismatch
+Nowak2017-QAP-2,A003,off,medium,source_scope_mismatch
+Nowak2017-QAP-2,A004,C0,medium,source_scope_mismatch
+Nowak2017-QAP-3,A001,off,medium,source_scope_mismatch
+Nowak2017-QAP-3,A002,off,medium,source_scope_mismatch
+Nowak2017-QAP-3,A003,C0,medium,source_scope_mismatch
+Nowak2017-QAP-3,A004,off,medium,source_scope_mismatch
+Mandi2024-DFL-1,A001,C0,high,agreement
+Mandi2024-DFL-1,A002,C0,high,agreement
+Mandi2024-DFL-1,A003,C0,high,agreement
+Mandi2024-DFL-1,A004,C0,high,agreement
+Mandi2024-DFL-2,A001,C1,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-2,A002,C0,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-2,A003,C0,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-2,A004,C0,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-3,A001,C0,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-3,A002,C0,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-3,A003,C1,medium,dynamic_or_deployment_escalation
+Mandi2024-DFL-3,A004,C0,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-1,A001,C0,high,agreement
+Elmachtoub2022-SPO-1,A002,C0,high,agreement
+Elmachtoub2022-SPO-1,A003,C0,high,agreement
+Elmachtoub2022-SPO-1,A004,C0,high,agreement
+Elmachtoub2022-SPO-2,A001,C1,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-2,A002,C0,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-2,A003,C0,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-2,A004,C0,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-3,A001,C0,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-3,A002,C0,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-3,A003,C1,medium,dynamic_or_deployment_escalation
+Elmachtoub2022-SPO-3,A004,C0,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-1,A001,C0,high,agreement
+Wilder2019-Melding-1,A002,C0,high,agreement
+Wilder2019-Melding-1,A003,C0,high,agreement
+Wilder2019-Melding-1,A004,C0,high,agreement
+Wilder2019-Melding-2,A001,C1,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-2,A002,C0,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-2,A003,C0,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-2,A004,C0,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-3,A001,C0,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-3,A002,C0,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-3,A003,C1,medium,dynamic_or_deployment_escalation
+Wilder2019-Melding-3,A004,C0,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-1,A001,C0,high,agreement
+Berthet2020-Perturb-1,A002,C0,high,agreement
+Berthet2020-Perturb-1,A003,C0,high,agreement
+Berthet2020-Perturb-1,A004,C0,high,agreement
+Berthet2020-Perturb-2,A001,C1,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-2,A002,C0,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-2,A003,C0,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-2,A004,C0,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-3,A001,C0,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-3,A002,C0,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-3,A003,C1,medium,dynamic_or_deployment_escalation
+Berthet2020-Perturb-3,A004,C0,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-1,A001,C0,high,agreement
+Poganvcic2020-BB-1,A002,C0,high,agreement
+Poganvcic2020-BB-1,A003,C0,high,agreement
+Poganvcic2020-BB-1,A004,C0,high,agreement
+Poganvcic2020-BB-2,A001,C1,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-2,A002,C0,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-2,A003,C0,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-2,A004,C0,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-3,A001,C0,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-3,A002,C0,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-3,A003,C1,medium,dynamic_or_deployment_escalation
+Poganvcic2020-BB-3,A004,C0,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-1,A001,C0,high,agreement
+Niepert2021-Implicit-1,A002,C0,high,agreement
+Niepert2021-Implicit-1,A003,C0,high,agreement
+Niepert2021-Implicit-1,A004,C0,high,agreement
+Niepert2021-Implicit-2,A001,C1,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-2,A002,C0,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-2,A003,C0,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-2,A004,C0,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-3,A001,C0,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-3,A002,C0,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-3,A003,C1,medium,dynamic_or_deployment_escalation
+Niepert2021-Implicit-3,A004,C0,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-1,A001,C0,high,agreement
+Petersen2024-Newton-1,A002,C0,high,agreement
+Petersen2024-Newton-1,A003,C0,high,agreement
+Petersen2024-Newton-1,A004,C0,high,agreement
+Petersen2024-Newton-2,A001,C1,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-2,A002,C0,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-2,A003,C0,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-2,A004,C0,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-3,A001,C0,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-3,A002,C0,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-3,A003,C1,medium,dynamic_or_deployment_escalation
+Petersen2024-Newton-3,A004,C0,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-1,A001,C0,high,agreement
+Lahoud2024-DataSP-1,A002,C0,high,agreement
+Lahoud2024-DataSP-1,A003,C0,high,agreement
+Lahoud2024-DataSP-1,A004,C0,high,agreement
+Lahoud2024-DataSP-2,A001,C1,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-2,A002,C0,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-2,A003,C0,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-2,A004,C0,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-3,A001,C0,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-3,A002,C0,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-3,A003,C1,medium,dynamic_or_deployment_escalation
+Lahoud2024-DataSP-3,A004,C0,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-1,A001,C3,high,agreement
+Rydin2026-GMS-1,A002,C3,high,agreement
+Rydin2026-GMS-1,A003,C3,high,agreement
+Rydin2026-GMS-1,A004,C3,high,agreement
+Rydin2026-GMS-2,A001,C4,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-2,A002,C3,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-2,A003,C3,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-2,A004,C3,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-3,A001,C3,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-3,A002,C2,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-3,A003,C3,medium,dynamic_or_deployment_escalation
+Rydin2026-GMS-3,A004,C3,medium,dynamic_or_deployment_escalation
+FrontierCO2026-1,A001,off,high,agreement
+FrontierCO2026-1,A002,off,high,agreement
+FrontierCO2026-1,A003,off,high,agreement
+FrontierCO2026-1,A004,off,high,agreement
+FrontierCO2026-2,A001,off,medium,source_scope_mismatch
+FrontierCO2026-2,A002,off,medium,source_scope_mismatch
+FrontierCO2026-2,A003,off,medium,source_scope_mismatch
+FrontierCO2026-2,A004,C0,medium,source_scope_mismatch
+FrontierCO2026-3,A001,off,medium,source_scope_mismatch
+FrontierCO2026-3,A002,off,medium,source_scope_mismatch
+FrontierCO2026-3,A003,C0,medium,source_scope_mismatch
+FrontierCO2026-3,A004,off,medium,source_scope_mismatch
+COBench2025-1,A001,off,high,agreement
+COBench2025-1,A002,off,high,agreement
+COBench2025-1,A003,off,high,agreement
+COBench2025-1,A004,off,high,agreement
+COBench2025-2,A001,off,medium,source_scope_mismatch
+COBench2025-2,A002,off,medium,source_scope_mismatch
+COBench2025-2,A003,off,medium,source_scope_mismatch
+COBench2025-2,A004,C0,medium,source_scope_mismatch
+COBench2025-3,A001,off,medium,source_scope_mismatch
+COBench2025-3,A002,off,medium,source_scope_mismatch
+COBench2025-3,A003,C0,medium,source_scope_mismatch
+COBench2025-3,A004,off,medium,source_scope_mismatch
+Deb2002-NSGAII-1,A001,C3,high,agreement
+Deb2002-NSGAII-1,A002,C3,high,agreement
+Deb2002-NSGAII-1,A003,C3,high,agreement
+Deb2002-NSGAII-1,A004,C3,high,agreement
+Deb2002-NSGAII-2,A001,C4,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-2,A002,C3,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-2,A003,C3,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-2,A004,C3,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-3,A001,C3,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-3,A002,C2,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-3,A003,C3,medium,dynamic_or_deployment_escalation
+Deb2002-NSGAII-3,A004,C3,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-1,A001,C3,high,agreement
+Zitzler1999-HV-1,A002,C3,high,agreement
+Zitzler1999-HV-1,A003,C3,high,agreement
+Zitzler1999-HV-1,A004,C3,high,agreement
+Zitzler1999-HV-2,A001,C4,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-2,A002,C3,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-2,A003,C3,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-2,A004,C3,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-3,A001,C3,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-3,A002,C2,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-3,A003,C3,medium,dynamic_or_deployment_escalation
+Zitzler1999-HV-3,A004,C3,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-1,A001,C3,high,agreement
+Knowles2006-HV-1,A002,C3,high,agreement
+Knowles2006-HV-1,A003,C3,high,agreement
+Knowles2006-HV-1,A004,C3,high,agreement
+Knowles2006-HV-2,A001,C4,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-2,A002,C3,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-2,A003,C3,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-2,A004,C3,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-3,A001,C3,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-3,A002,C2,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-3,A003,C3,medium,dynamic_or_deployment_escalation
+Knowles2006-HV-3,A004,C3,medium,dynamic_or_deployment_escalation
+HELM2023-1,A001,off,high,agreement
+HELM2023-1,A002,off,high,agreement
+HELM2023-1,A003,off,high,agreement
+HELM2023-1,A004,off,high,agreement
+HELM2023-2,A001,off,medium,source_scope_mismatch
+HELM2023-2,A002,off,medium,source_scope_mismatch
+HELM2023-2,A003,off,medium,source_scope_mismatch
+HELM2023-2,A004,C0,medium,source_scope_mismatch
+HELM2023-3,A001,off,medium,source_scope_mismatch
+HELM2023-3,A002,off,medium,source_scope_mismatch
+HELM2023-3,A003,C0,medium,source_scope_mismatch
+HELM2023-3,A004,off,medium,source_scope_mismatch
+WILDS2021-1,A001,C5,high,agreement
+WILDS2021-1,A002,C5,high,agreement
+WILDS2021-1,A003,C5,high,agreement
+WILDS2021-1,A004,C5,high,agreement
+WILDS2021-2,A001,C6,medium,dynamic_or_deployment_escalation
+WILDS2021-2,A002,C5,medium,dynamic_or_deployment_escalation
+WILDS2021-2,A003,C5,medium,dynamic_or_deployment_escalation
+WILDS2021-2,A004,C4,medium,dynamic_or_deployment_escalation
+WILDS2021-3,A001,C5,medium,dynamic_or_deployment_escalation
+WILDS2021-3,A002,C4,medium,dynamic_or_deployment_escalation
+WILDS2021-3,A003,C5,medium,dynamic_or_deployment_escalation
+WILDS2021-3,A004,C5,medium,dynamic_or_deployment_escalation
+CheckList2020-1,A001,off,high,agreement
+CheckList2020-1,A002,off,high,agreement
+CheckList2020-1,A003,off,high,agreement
+CheckList2020-1,A004,off,high,agreement
+CheckList2020-2,A001,off,medium,source_scope_mismatch
+CheckList2020-2,A002,off,medium,source_scope_mismatch
+CheckList2020-2,A003,off,medium,source_scope_mismatch
+CheckList2020-2,A004,C0,medium,source_scope_mismatch
+CheckList2020-3,A001,off,medium,source_scope_mismatch
+CheckList2020-3,A002,off,medium,source_scope_mismatch
+CheckList2020-3,A003,C0,medium,source_scope_mismatch
+CheckList2020-3,A004,off,medium,source_scope_mismatch
+Dynabench2021-1,A001,C5,high,agreement
+Dynabench2021-1,A002,C5,high,agreement
+Dynabench2021-1,A003,C5,high,agreement
+Dynabench2021-1,A004,C5,high,agreement
+Dynabench2021-2,A001,C6,medium,dynamic_or_deployment_escalation
+Dynabench2021-2,A002,C5,medium,dynamic_or_deployment_escalation
+Dynabench2021-2,A003,C5,medium,dynamic_or_deployment_escalation
+Dynabench2021-2,A004,C4,medium,dynamic_or_deployment_escalation
+Dynabench2021-3,A001,C5,medium,dynamic_or_deployment_escalation
+Dynabench2021-3,A002,C4,medium,dynamic_or_deployment_escalation
+Dynabench2021-3,A003,C5,medium,dynamic_or_deployment_escalation
+Dynabench2021-3,A004,C5,medium,dynamic_or_deployment_escalation
+Datasheets2021-1,A001,off,high,agreement
+Datasheets2021-1,A002,off,high,agreement
+Datasheets2021-1,A003,off,high,agreement
+Datasheets2021-1,A004,off,high,agreement
+Datasheets2021-2,A001,off,medium,source_scope_mismatch
+Datasheets2021-2,A002,off,medium,source_scope_mismatch
+Datasheets2021-2,A003,off,medium,source_scope_mismatch
+Datasheets2021-2,A004,C0,medium,source_scope_mismatch
+Datasheets2021-3,A001,off,medium,source_scope_mismatch
+Datasheets2021-3,A002,off,medium,source_scope_mismatch
+Datasheets2021-3,A003,C0,medium,source_scope_mismatch
+Datasheets2021-3,A004,off,medium,source_scope_mismatch
+ModelCards2019-1,A001,off,high,agreement
+ModelCards2019-1,A002,off,high,agreement
+ModelCards2019-1,A003,off,high,agreement
+ModelCards2019-1,A004,off,high,agreement
+ModelCards2019-2,A001,off,medium,source_scope_mismatch
+ModelCards2019-2,A002,off,medium,source_scope_mismatch
+ModelCards2019-2,A003,off,medium,source_scope_mismatch
+ModelCards2019-2,A004,C0,medium,source_scope_mismatch
+ModelCards2019-3,A001,off,medium,source_scope_mismatch
+ModelCards2019-3,A002,off,medium,source_scope_mismatch
+ModelCards2019-3,A003,C0,medium,source_scope_mismatch
+ModelCards2019-3,A004,off,medium,source_scope_mismatch
+BenchmarkCards2025-1,A001,off,high,agreement
+BenchmarkCards2025-1,A002,off,high,agreement
+BenchmarkCards2025-1,A003,off,high,agreement
+BenchmarkCards2025-1,A004,off,high,agreement
+BenchmarkCards2025-2,A001,off,medium,source_scope_mismatch
+BenchmarkCards2025-2,A002,off,medium,source_scope_mismatch
+BenchmarkCards2025-2,A003,off,medium,source_scope_mismatch
+BenchmarkCards2025-2,A004,C0,medium,source_scope_mismatch
+BenchmarkCards2025-3,A001,off,medium,source_scope_mismatch
+BenchmarkCards2025-3,A002,off,medium,source_scope_mismatch
+BenchmarkCards2025-3,A003,C0,medium,source_scope_mismatch
+BenchmarkCards2025-3,A004,off,medium,source_scope_mismatch
+ReproChecklist2021-1,A001,off,high,agreement
+ReproChecklist2021-1,A002,off,high,agreement
+ReproChecklist2021-1,A003,off,high,agreement
+ReproChecklist2021-1,A004,off,high,agreement
+ReproChecklist2021-2,A001,off,medium,source_scope_mismatch
+ReproChecklist2021-2,A002,off,medium,source_scope_mismatch
+ReproChecklist2021-2,A003,off,medium,source_scope_mismatch
+ReproChecklist2021-2,A004,C0,medium,source_scope_mismatch
+ReproChecklist2021-3,A001,off,medium,source_scope_mismatch
+ReproChecklist2021-3,A002,off,medium,source_scope_mismatch
+ReproChecklist2021-3,A003,C0,medium,source_scope_mismatch
+ReproChecklist2021-3,A004,off,medium,source_scope_mismatch
+Croissant2024-1,A001,off,high,agreement
+Croissant2024-1,A002,off,high,agreement
+Croissant2024-1,A003,off,high,agreement
+Croissant2024-1,A004,off,high,agreement
+Croissant2024-2,A001,off,medium,source_scope_mismatch
+Croissant2024-2,A002,off,medium,source_scope_mismatch
+Croissant2024-2,A003,off,medium,source_scope_mismatch
+Croissant2024-2,A004,C0,medium,source_scope_mismatch
+Croissant2024-3,A001,off,medium,source_scope_mismatch
+Croissant2024-3,A002,off,medium,source_scope_mismatch
+Croissant2024-3,A003,C0,medium,source_scope_mismatch
+Croissant2024-3,A004,off,medium,source_scope_mismatch
+VRP-LKH-1,A001,C0,high,agreement
+VRP-LKH-1,A002,C0,high,agreement
+VRP-LKH-1,A003,C0,high,agreement
+VRP-LKH-1,A004,C0,high,agreement
+VRP-LKH-2,A001,C1,medium,dynamic_or_deployment_escalation
+VRP-LKH-2,A002,C0,medium,dynamic_or_deployment_escalation
+VRP-LKH-2,A003,C0,medium,dynamic_or_deployment_escalation
+VRP-LKH-2,A004,C0,medium,dynamic_or_deployment_escalation
+VRP-LKH-3,A001,C0,medium,dynamic_or_deployment_escalation
+VRP-LKH-3,A002,C0,medium,dynamic_or_deployment_escalation
+VRP-LKH-3,A003,C1,medium,dynamic_or_deployment_escalation
+VRP-LKH-3,A004,C0,medium,dynamic_or_deployment_escalation
+OR-Tools-1,A001,C0,high,agreement
+OR-Tools-1,A002,C0,high,agreement
+OR-Tools-1,A003,C0,high,agreement
+OR-Tools-1,A004,C0,high,agreement
+OR-Tools-2,A001,C1,medium,dynamic_or_deployment_escalation
+OR-Tools-2,A002,C0,medium,dynamic_or_deployment_escalation
+OR-Tools-2,A003,C0,medium,dynamic_or_deployment_escalation
+OR-Tools-2,A004,C0,medium,dynamic_or_deployment_escalation
+OR-Tools-3,A001,C0,medium,dynamic_or_deployment_escalation
+OR-Tools-3,A002,C0,medium,dynamic_or_deployment_escalation
+OR-Tools-3,A003,C1,medium,dynamic_or_deployment_escalation
+OR-Tools-3,A004,C0,medium,dynamic_or_deployment_escalation
+Gurobi-CO-1,A001,C0,high,agreement
+Gurobi-CO-1,A002,C0,high,agreement
+Gurobi-CO-1,A003,C0,high,agreement
+Gurobi-CO-1,A004,C0,high,agreement
+Gurobi-CO-2,A001,C1,medium,dynamic_or_deployment_escalation
+Gurobi-CO-2,A002,C0,medium,dynamic_or_deployment_escalation
+Gurobi-CO-2,A003,C0,medium,dynamic_or_deployment_escalation
+Gurobi-CO-2,A004,C0,medium,dynamic_or_deployment_escalation
+Gurobi-CO-3,A001,C0,medium,dynamic_or_deployment_escalation
+Gurobi-CO-3,A002,C0,medium,dynamic_or_deployment_escalation
+Gurobi-CO-3,A003,C1,medium,dynamic_or_deployment_escalation
+Gurobi-CO-3,A004,C0,medium,dynamic_or_deployment_escalation

real_source_blinding_map.csv

@@ -0,0 +1,41 @@
+anon_source_id,source_hash_12,reviewer_auditable_source_id,source_title,source_family,source_scope,claim_count,note
+SRC-001,8bedbcb4c784,Bello2017-NCO,Neural Combinatorial Optimization with Reinforcement Learning,neural CO,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-002,7fb929f4e197,BenchmarkCards2025,BenchmarkCards,benchmark documentation,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-003,bf3bc17d056e,Berthet2020-Perturb,Learning with Differentiable Perturbed Optimizers,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-004,492c9f59ee43,COBench2025,CO-Bench,benchmarking,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-005,caf7b5f40ee7,CheckList2020,CheckList,behavioral testing,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-006,be99aca66589,ChenTian2019-NR,Learning to Perform Local Rewriting for Combinatorial Optimization,neural rewriting,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-007,69caa53170a7,Croissant2024,Croissant Metadata,metadata,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-008,9e627348f556,Dai2017-S2V,Learning Combinatorial Optimization Algorithms over Graphs,graph neural CO,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-009,21c8aefdad79,Datasheets2021,Datasheets for Datasets,documentation,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-010,9db8ad3677c7,Deb2002-NSGAII,NSGA-II,multiobjective,preference_or_multiobjective_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-011,d1b4cf104f5a,Dynabench2021,Dynabench,dynamic benchmarks,dynamic_stress_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-012,20ebaa6bdeab,Elmachtoub2022-SPO,Smart Predict-then-Optimize,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-013,d9626b64c635,FrontierCO2026,FrontierCO,large-scale ML4CO,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-014,ab8a2ed8f356,Fu2021-GeneralizeTSP,Generalize a Small Pre-trained Model to Large TSP,TSP transfer,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-015,ad37b768301d,Gurobi-CO,Mathematical Programming Solver Baselines,solver interface,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-016,74348d4f83a5,HELM2023,Holistic Evaluation of Language Models,evaluation science,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-017,d15bf1d23c44,Joshi2019-GCN-TSP,An Efficient Graph Convolutional Network Technique for TSP,graph neural TSP,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-018,381380a27377,Khalil2017-L2S,Learning to Run Heuristics in CO,learning to search,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-019,35d5a8f027ab,Kim2022-SymNCO,Symmetric Neural Combinatorial Optimization,neural routing,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-020,c128692a4fd5,Knowles2006-HV,Hypervolume Indicator Tutorial,multiobjective,preference_or_multiobjective_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-021,33070a3198b3,Kool2019-AM,"Attention, Learn to Solve Routing Problems",neural routing,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-022,01f2c467a09c,Lahoud2024-DataSP,DataSP,shortest path,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-023,4f0f7e8c48a1,Ma2021-NeuroLKH,NeuroLKH,solver guidance,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-024,158fb4890e8b,Mandi2024-DFL,Decision-Focused Learning benchmark,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-025,27b563897122,ModelCards2019,Model Cards,documentation,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-026,79c028a35351,Nazari2018-VRP,Reinforcement Learning for Solving the Vehicle Routing Problem,dynamic VRP,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-027,cc0bb2ec99d5,Niepert2021-Implicit,Implicit MLE for Combinatorial Optimization,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-028,f9832cb26cba,Nowak2017-QAP,A Note on Learning Algorithms for QAP,graph neural CO,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-029,6e8396622af7,OR-Tools,OR-Tools Routing,classical solver,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-030,bdbae8809853,Petersen2024-Newton,Newtons Method for Differentiable CO Layers,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-031,2cf7745be99b,Poganvcic2020-BB,Differentiation of Blackbox Combinatorial Solvers,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-032,0c8f5c96d8b4,Qiu2022-DIMES,DIMES,differentiable CO,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-033,a6bf8ed6948c,ReproChecklist2021,Improving Reproducibility in ML Research,reproducibility,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-034,527e3468fd16,Rydin2026-GMS,Beyond Simple Graphs / GMS,multiobjective multigraph,preference_or_multiobjective_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-035,029c20b8644e,VRP-LKH,LKH for Routing,classical solver,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-036,7ed4b0da80da,Vinyals2015-Pointer,Pointer Networks,neural sequence routing,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-037,9e0551040233,WILDS2021,WILDS,distribution shift,dynamic_stress_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-038,6881b6d4e540,Wilder2019-Melding,Melding the Data-Decisions Pipeline,decision-focused,scalar_or_bridge_decision_quality,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-039,dc700c5d9585,Xin2021-POMO,Policy Optimization with Multiple Optima for Reinforcement Learning,neural routing,off_contract_or_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.
+SRC-040,147a06c82ec0,Zitzler1999-HV,Multiobjective Evolutionary Algorithms and Hypervolume,multiobjective,preference_or_multiobjective_analogue,3,Source identifiers are anonymized for tables but include canonical source IDs/titles for reviewer audit.