Add 31 PRBench task specifications (instructions, task.yaml, paper markdown; ground truth withheld)
a48f2b1 verified | # PRBench Task Configuration | |
| task_id: task_bude_2021 | |
| # --- Paper Info --- | |
| paper: | |
| title: "Accelerating simulations of electromagnetic waves in hot, magnetized fusion plasmas" | |
| author: "R H S Budé, D Van Eester, J van Dijk, R J E Jaspers and A B Smolders" | |
| doi: "10.1088/1361-6587/abd619" | |
| year: 2021 | |
| paper_file: "Bude2021.md" | |
| # --- Files --- | |
| instruction_file: "instruction.md" | |
| metadata_file: "metadata.md" | |
| ground_truth_data_dir: "data" | |
| # --- Expected Outputs --- | |
| expected_outputs: | |
| analysis: | |
| - "reproduction/ANALYSIS.md" | |
| code: | |
| - "reproduction/plasma.py" | |
| - "reproduction/dielectric.py" | |
| - "reproduction/solvers.py" | |
| - "reproduction/run_sim.py" | |
| - "reproduction/fig2_profiles.py" | |
| - "reproduction/fig3_4_tensor.py" | |
| - "reproduction/fig5_fields.py" | |
| - "reproduction/fig6_power.py" | |
| - "reproduction/fig7_convergence.py" | |
| - "reproduction/fig8_timing.py" | |
| - "reproduction/fig9_params.py" | |
| - "reproduction/fig10_approx.py" | |
| - "reproduction/table1_scaling.py" | |
| data: | |
| - "data/fig2_profiles.csv" | |
| - "data/fig3_4_tensor_fit.csv" | |
| - "data/fig5_efield.csv" | |
| - "data/fig6_power.csv" | |
| - "data/fig7_error_scaling.csv" | |
| - "data/fig8_timing.csv" | |
| - "data/fig9_param_sweep.csv" | |
| - "data/fig10_function_approx.csv" | |
| - "data/table1_extrapolation.csv" | |
| # --- Docker Environment --- | |
| docker: | |
| image: "python:3.11-slim" | |
| memory_limit: "8g" | |
| timeout: 14400 # 4 hours max | |
| pip_install: | |
| - numpy | |
| - scipy | |
| - matplotlib | |
| # --- Grading Rubric --- | |
| grading: | |
| dimensions: | |
| - name: "methodology_understanding" | |
| weight: 0.20 | |
| description: > | |
| Does the agent's ANALYSIS.md correctly describe the electromagnetic wave | |
| simulation methodology, including the three solver models (Truncated Taylor, | |
| Truncated Polynomial Fit, All-Orders), dielectric tensor construction, | |
| polynomial fitting procedure, finite difference stencils, boundary conditions, | |
| and measurement procedures? Compare against metadata.md. | |
| - name: "code_correctness" | |
| weight: 0.15 | |
| description: > | |
| Are all required code files present? Does the code implement the solvers | |
| correctly: plasma profiles, hot plasma dielectric tensor, Taylor and | |
| Polynomial Fit differential operators, sparse matrix assembly, boundary | |
| conditions, power absorption, and RRSE error metric? | |
| - name: "data_accuracy" | |
| weight: 0.50 | |
| description: > | |
| How well do the generated CSV data files match the ground truth reference data? | |
| Check column names, row counts, and value ranges. Tolerance ~5-10% for | |
| field solutions; profiles and function approximations should match closely. | |
| - name: "completeness" | |
| weight: 0.15 | |
| description: > | |
| Are all 9 CSV files, all 13 code files, and the ANALYSIS.md present? | |
| Are the files non-trivial (not empty/placeholder)? | |
| # --- Code-Only Evaluation Mode --- | |
| code_only: | |
| instruction_suffix: | | |
| ## IMPORTANT: Code-Only Mode | |
| In this task you are in **code-only mode**. You must: | |
| 1. Read and analyze the paper | |
| 2. Write `reproduction/ANALYSIS.md` with your methodology analysis | |
| 3. Implement ALL code files under `reproduction/` | |
| 4. **DO NOT** run the simulations or generate CSV data files | |
| 5. Focus on writing correct, complete, and well-structured code | |
| Your deliverables are ONLY the analysis document and code files. | |
| Do NOT produce any CSV data files. | |
| expected_outputs: | |
| analysis: | |
| - "reproduction/ANALYSIS.md" | |
| code: | |
| - "reproduction/plasma.py" | |
| - "reproduction/dielectric.py" | |
| - "reproduction/solvers.py" | |
| - "reproduction/run_sim.py" | |
| - "reproduction/fig2_profiles.py" | |
| - "reproduction/fig3_4_tensor.py" | |
| - "reproduction/fig5_fields.py" | |
| - "reproduction/fig6_power.py" | |
| - "reproduction/fig7_convergence.py" | |
| - "reproduction/fig8_timing.py" | |
| - "reproduction/fig9_params.py" | |
| - "reproduction/fig10_approx.py" | |
| - "reproduction/table1_scaling.py" | |
| grading: | |
| dimensions: | |
| - name: "methodology_understanding" | |
| weight: 0.25 | |
| description: > | |
| Does ANALYSIS.md correctly describe the electromagnetic wave simulation | |
| methodology, including the three solver models, dielectric tensor, | |
| polynomial fitting, finite difference stencils, and boundary conditions? | |
| - name: "code_correctness" | |
| weight: 0.50 | |
| description: > | |
| Read every code file carefully. Check whether the implementation is | |
| physically and algorithmically correct: | |
| - Plasma profiles (density, temperature, magnetic field) | |
| - Hot plasma dielectric tensor (K0-K5 terms, Bessel functions, Plasma Dispersion Function) | |
| - Truncated Taylor solver (2nd order expansion, 3-point stencil) | |
| - Truncated Polynomial Fit solver (arbitrary order fitting, Fornberg stencils) | |
| - Sparse matrix assembly and boundary conditions | |
| - Power absorption calculation | |
| - RRSE error metric | |
| - Computational timing and scaling extrapolation | |
| Compare against ground truth formulas in metadata. Flag any bugs | |
| that would produce incorrect results. | |
| - name: "code_quality" | |
| weight: 0.10 | |
| description: > | |
| Is the code well-structured, readable, and does it follow the | |
| expected file organization? | |
| - name: "completeness" | |
| weight: 0.15 | |
| description: > | |
| Are all 13 code files and the ANALYSIS.md present and non-trivial? | |