| 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 | |