UnipatAI Claude Opus 5.5 commited on
Commit
ae5b1ef
·
verified ·
1 Parent(s): 97c654b

Release v1.1: remove tied items, neutral item order, Spearman-only grader

Browse files

Same 100 tasks, 489 groups and measured values as v1.0 (tag v1.0, used in the paper). Removes 247 tied items in 52 groups, regenerates item IDs and lists items in ID order, drops the auxiliary score, cleans source identifiers from public files, moves task.toml to Harbor schema 1.4, and keeps the picalm donor-field removal from v1.0. README gains a Versions section.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +14 -5
  2. SHA256SUMS +0 -0
  3. task_cards.parquet +2 -2
  4. tasks/agcute-bending-endpoints-ranking/environment/public/contract.json +33 -43
  5. tasks/agcute-bending-endpoints-ranking/environment/public/experiment.md +1 -1
  6. tasks/agcute-bending-endpoints-ranking/environment/public/information_policy.md +2 -0
  7. tasks/agcute-bending-endpoints-ranking/environment/public/protocol.json +58 -58
  8. tasks/agcute-bending-endpoints-ranking/environment/public/targets.csv +18 -18
  9. tasks/agcute-bending-endpoints-ranking/instruction.md +57 -70
  10. tasks/agcute-bending-endpoints-ranking/solution/solve.sh +21 -21
  11. tasks/agcute-bending-endpoints-ranking/task.toml +4 -6
  12. tasks/agcute-bending-endpoints-ranking/tests/answers.json +55 -837
  13. tasks/agcute-bending-endpoints-ranking/tests/contract.json +33 -43
  14. tasks/agcute-bending-endpoints-ranking/tests/grade.py +207 -97
  15. tasks/agcute-bending-endpoints-ranking/tests/measurements.csv +18 -18
  16. tasks/agcute-bending-endpoints-ranking/tests/original_grade.py +0 -423
  17. tasks/agcute-bending-endpoints-ranking/tests/random_baseline.json +0 -66
  18. tasks/agcute-bending-endpoints-ranking/tests/scoring_hashes.json +5 -0
  19. tasks/alkene-transfer-conditions-ranking/environment/public/contract.json +31 -44
  20. tasks/alkene-transfer-conditions-ranking/environment/public/experiment.md +8 -8
  21. tasks/alkene-transfer-conditions-ranking/environment/public/information_policy.md +2 -0
  22. tasks/alkene-transfer-conditions-ranking/environment/public/protocol.json +70 -85
  23. tasks/alkene-transfer-conditions-ranking/environment/public/targets.csv +15 -16
  24. tasks/alkene-transfer-conditions-ranking/instruction.md +55 -72
  25. tasks/alkene-transfer-conditions-ranking/solution/solve.sh +18 -19
  26. tasks/alkene-transfer-conditions-ranking/task.toml +4 -6
  27. tasks/alkene-transfer-conditions-ranking/tests/answers.json +42 -560
  28. tasks/alkene-transfer-conditions-ranking/tests/contract.json +31 -44
  29. tasks/alkene-transfer-conditions-ranking/tests/grade.py +207 -97
  30. tasks/alkene-transfer-conditions-ranking/tests/measurements.csv +15 -16
  31. tasks/alkene-transfer-conditions-ranking/tests/original_grade.py +0 -423
  32. tasks/alkene-transfer-conditions-ranking/tests/random_baseline.json +0 -67
  33. tasks/alkene-transfer-conditions-ranking/tests/scoring_hashes.json +5 -0
  34. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/contract.json +18 -26
  35. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/experiment.md +4 -4
  36. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/information_policy.md +2 -0
  37. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/protocol.json +12 -15
  38. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/targets.csv +4 -5
  39. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/instruction.md +30 -42
  40. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/solution/solve.sh +3 -4
  41. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/task.toml +4 -6
  42. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/answers.json +21 -419
  43. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/contract.json +18 -26
  44. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/grade.py +207 -97
  45. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/measurements.csv +4 -5
  46. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/original_grade.py +0 -451
  47. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/random_baseline.json +0 -54
  48. tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/scoring_hashes.json +5 -0
  49. tasks/alkyne-bridged-ester-palladium-screen/environment/public/contract.json +17 -21
  50. tasks/alkyne-bridged-ester-palladium-screen/environment/public/experiment.md +1 -1
README.md CHANGED
@@ -62,7 +62,7 @@ from datasets import load_dataset
62
  cards = load_dataset("UnipatAI/SPU", split="test")
63
  ```
64
 
65
- Run a task with Harbor 0.25.0 (Python 3.12 or later) and Docker. This oracle check replays the reference answer and needs no GPU:
66
 
67
  ```bash
68
  hf download UnipatAI/SPU --repo-type dataset --local-dir SPU && cd SPU
@@ -74,21 +74,30 @@ In the paper, each run had 8 CPU cores, 16 GiB of memory, one NVIDIA H20 GPU, we
74
 
75
  ## Scoring
76
 
77
- Each task has one or more comparison groups. The grader computes the Spearman correlation between the submitted ranking and the measured values in each group and combines the groups with the fixed weights given in `instruction.md`. A perfect ranking scores 1 and random rankings average 0. Tied measurements get average ranks, so on 33 tasks the best possible score is slightly below 1. A missing or malformed submission scores −1, the same as an exactly reversed ranking: check `submission_status` in the grader's `report.json`.
78
 
79
  The paper's headline metric, the task-adjusted mean, is fitted to the scores of all thirteen evaluated configurations and cannot be computed from a single run. When reporting results on these tasks, give the mean over scored tasks, the number of scored tasks, the model version, reasoning effort and harness. The paper's audited web tools and source-identifier screening are not part of this release, so results obtained without them are not directly comparable to the paper's.
80
 
 
 
 
 
 
 
 
 
 
 
 
 
81
  ## Sources
82
 
83
  Each task is built from one published paper, cited in `SOURCES.json`. Some biology and medicine tasks include published context about human-derived samples or cell lines from the source studies.
84
 
85
  ## Known issues
86
 
87
- - In `mem-cyclopropene-phosphine-conditions-ranking`, one `catalyst_scaffold` value in `protocol.json` joins the image path and a note (`structures/mem_catalyst.pngwithdrawnstereochemistry`). The image is `environment/public/structures/mem_catalyst.png`, and the experiment description links it correctly.
88
  - In `picalm-lipid-class-abundance`, the donor sex, age, ancestry, APOE genotype and cell-bank IDs given in the paper's runs were removed from `cell_context` in `protocol.json` and `experiment.md`. Donor codes, conditions, measurements and grading are unchanged.
89
 
90
- All other task files are as they were run in the paper.
91
-
92
  ## Authors
93
 
94
  **Project Team:** Zengji Tu<sup>1*</sup>, Xiang Liu<sup>1*</sup>, Jiarui Sun<sup>1</sup>, Ruoyu Wu<sup>1</sup>, Zhihui Qi<sup>1</sup>, Chengda Wu<sup>1</sup>, Yutao Zhou<sup>1</sup>, Bo Yang<sup>1</sup>, Ziyan Zhang<sup>1</sup>, Jialong Wu<sup>1</sup>, Liang Chen<sup>1†‡</sup>, Kuan Li<sup>1†</sup>
 
62
  cards = load_dataset("UnipatAI/SPU", split="test")
63
  ```
64
 
65
+ Run a task with Harbor 0.24.0 or later (Python 3.12 or later) and Docker. This oracle check replays the reference answer and needs no GPU:
66
 
67
  ```bash
68
  hf download UnipatAI/SPU --repo-type dataset --local-dir SPU && cd SPU
 
74
 
75
  ## Scoring
76
 
77
+ Each task has one or more comparison groups. The grader computes the Spearman correlation between the submitted ranking and the measured values in each group and combines the groups with the fixed weights given in `instruction.md`. A perfect ranking scores 1 and random rankings average 0. No group contains tied measurements, so every task can reach 1. A missing or malformed submission scores −1, the same as an exactly reversed ranking: check `submission_status` in the grader's `report.json`. If an agent stops under the information policy, it writes `SPU_POLICY_STOP` on the first line of `report.md`; the grader records this separately.
78
 
79
  The paper's headline metric, the task-adjusted mean, is fitted to the scores of all thirteen evaluated configurations and cannot be computed from a single run. When reporting results on these tasks, give the mean over scored tasks, the number of scored tasks, the model version, reasoning effort and harness. The paper's audited web tools and source-identifier screening are not part of this release, so results obtained without them are not directly comparable to the paper's.
80
 
81
+ ## Versions
82
+
83
+ This is version 1.1 (`task_version = "1.1.0"`). The paper's results were obtained on version 1.0; load it with `revision="v1.0"`. Version 1.1 keeps the same 100 tasks, 489 comparison groups and measured values, and changes the following:
84
+
85
+ - **Tied items removed.** In 52 groups of 33 tasks, items whose measured value equalled that of another item in the same group were removed, so the release has 3,700 items instead of 3,947. The 93 structure images of removed items in three tasks were removed with them, and the item counts in the task descriptions were updated.
86
+ - **Neutral item order.** Item IDs were regenerated and items are listed in ID order, so the order of `targets.csv` and of the submission example no longer correlates with the measured values.
87
+ - **Spearman only.** The auxiliary pair-accuracy score and its files (`random_baseline.json`, `original_grade.py`) were removed. `tests/scoring_hashes.json` binds the grader to its answers and contract.
88
+ - **Cleaner inputs.** In ten tasks, source figure, table and compound numbers, research-group identifiers and evaluation-process wording were removed from the public files, a broken image path in `mem-cyclopropene-phosphine-conditions-ranking` was fixed, and unrounded composition values in `hydrogel-additional-formulation-adhesion` were rounded to three decimals. Empty source-limitation sections and missing protocol fields were filled in, and spacing in descriptions was repaired.
89
+ - **Harbor schema 1.4.** `task.toml` uses `network_mode` instead of `allow_internet`.
90
+
91
+ Scores on version 1.1 are not directly comparable to the paper's.
92
+
93
  ## Sources
94
 
95
  Each task is built from one published paper, cited in `SOURCES.json`. Some biology and medicine tasks include published context about human-derived samples or cell lines from the source studies.
96
 
97
  ## Known issues
98
 
 
99
  - In `picalm-lipid-class-abundance`, the donor sex, age, ancestry, APOE genotype and cell-bank IDs given in the paper's runs were removed from `cell_context` in `protocol.json` and `experiment.md`. Donor codes, conditions, measurements and grading are unchanged.
100
 
 
 
101
  ## Authors
102
 
103
  **Project Team:** Zengji Tu<sup>1*</sup>, Xiang Liu<sup>1*</sup>, Jiarui Sun<sup>1</sup>, Ruoyu Wu<sup>1</sup>, Zhihui Qi<sup>1</sup>, Chengda Wu<sup>1</sup>, Yutao Zhou<sup>1</sup>, Bo Yang<sup>1</sup>, Ziyan Zhang<sup>1</sup>, Jialong Wu<sup>1</sup>, Liang Chen<sup>1†‡</sup>, Kuan Li<sup>1†</sup>
SHA256SUMS CHANGED
The diff for this file is too large to render. See raw diff
 
task_cards.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ab80ef74a4226fedab9512372dcbee8f5b905e81350ff88b87fcd3b993f9e71b
3
- size 30582
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:181e7f58b704f7b721ec04dabbab423bb58a6a965ff066a410979c03635acafe
3
+ size 30211
tasks/agcute-bending-endpoints-ranking/environment/public/contract.json CHANGED
@@ -1,60 +1,50 @@
1
  {
2
  "task_id": "agcute-bending-endpoints-ranking",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "max_stress",
7
  "label": "Alloy composition series: bending stress",
8
  "item_ids": [
9
- "exp_65960918c5e1",
10
- "exp_0b361eedc092",
11
- "exp_0a801637748b",
12
- "exp_39fd1eed6b1c",
13
- "exp_365507fb236c",
14
- "exp_3c3c8522da9d",
15
- "exp_fc4984656462",
16
- "exp_3fed9f8f07c9",
17
- "exp_aaa9e2c6f5b9"
18
  ],
19
  "direction": "ascending",
20
  "group_weight": 0.5,
21
- "k": 2,
22
- "epsilon": 0,
23
- "uncertainty_model": "optional_reported_uncertainty",
24
- "difference_standard_error": "sem_sum_upper_bound",
25
  "endpoint": {
26
  "name": "Maximum bending stress",
27
  "unit": "MPa",
28
  "definition_en": "Largest recorded engineering bending stress in each curve."
29
- },
30
- "uncertainty_definition": "Reliable independent uncertainty is not available for this panel. Unequal reported values are compared directly; uncertainty is not assumed to be zero."
31
  },
32
  {
33
  "group_id": "strain_at_max",
34
  "label": "Alloy composition series: peak-stress strain",
35
  "item_ids": [
36
- "exp_54b8fc8b3571",
37
- "exp_3aa8fedf978b",
38
- "exp_2d9d2ad06f6e",
39
- "exp_8cfb0d14a259",
40
- "exp_d53afffea8ad",
41
- "exp_69a2a5342ec6",
42
- "exp_eb05cc2a9dc8",
43
- "exp_7b741589c10a",
44
- "exp_4369fc61d66d"
45
  ],
46
  "direction": "ascending",
47
  "group_weight": 0.5,
48
- "k": 2,
49
- "epsilon": 0,
50
- "uncertainty_model": "optional_reported_uncertainty",
51
- "difference_standard_error": "sem_sum_upper_bound",
52
  "endpoint": {
53
  "name": "Engineering strain at maximum bending stress",
54
  "unit": "%",
55
  "definition_en": "Engineering strain at the same sampled point as the largest bending stress, using the earliest occurrence if maxima repeat."
56
- },
57
- "uncertainty_definition": "Reliable independent uncertainty is not available for this panel. Unequal reported values are compared directly; uncertainty is not assumed to be zero."
58
  }
59
  ],
60
  "scoring_method": "spearman",
@@ -66,18 +56,9 @@
66
  "spearman_tie_method": "average",
67
  "spearman_undefined_policy": "task_error",
68
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
69
- "secondary_scoring_method": "equal_pair_inversion_random95",
70
- "ranking_reward_range": [
71
- 0,
72
- 1
73
- ],
74
- "random_quantile": 0.95,
75
- "unknown_uncertainty_policy": "use_reported_value",
76
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
77
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
78
  "format_errors": {
79
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
80
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
81
  },
82
  "max_submission_bytes": 1048576,
83
  "minimum_conditions_per_group": 2,
@@ -94,5 +75,14 @@
94
  "report_path": "/logs/artifacts/report.md",
95
  "sources_path": "/logs/artifacts/sources.json",
96
  "analysis_path": "/logs/artifacts/analysis/",
97
- "forbidden_result_scope": "Experimental bending stress-strain, flexural failure, toughness or equivalent mechanical rankings for Ag-Cu-Te-Se-S chalcogenide alloy composition and vacancy series are same-type results and prohibited, including related composition or test-protocol variants. General solid mechanics and independent chemical or crystallographic basic facts outside these result panels are allowed."
 
 
 
 
 
 
 
 
 
98
  }
 
1
  {
2
  "task_id": "agcute-bending-endpoints-ranking",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "max_stress",
7
  "label": "Alloy composition series: bending stress",
8
  "item_ids": [
9
+ "exp_003fe25c3f91",
10
+ "exp_224a125b3d5a",
11
+ "exp_63b9d2553a78",
12
+ "exp_672208100095",
13
+ "exp_7c4f7caec6cf",
14
+ "exp_c339123d0b13",
15
+ "exp_cf10a9056f26",
16
+ "exp_ea54530b4221",
17
+ "exp_ff5031900fc6"
18
  ],
19
  "direction": "ascending",
20
  "group_weight": 0.5,
 
 
 
 
21
  "endpoint": {
22
  "name": "Maximum bending stress",
23
  "unit": "MPa",
24
  "definition_en": "Largest recorded engineering bending stress in each curve."
25
+ }
 
26
  },
27
  {
28
  "group_id": "strain_at_max",
29
  "label": "Alloy composition series: peak-stress strain",
30
  "item_ids": [
31
+ "exp_0a7a7c7f2b75",
32
+ "exp_235fb0dc6cd6",
33
+ "exp_351f1bf17bc1",
34
+ "exp_3df219ed4754",
35
+ "exp_6cc95c89ecb5",
36
+ "exp_7ae22fdad3ab",
37
+ "exp_84f36130f06f",
38
+ "exp_d1a81b569dde",
39
+ "exp_fb7f19041d5a"
40
  ],
41
  "direction": "ascending",
42
  "group_weight": 0.5,
 
 
 
 
43
  "endpoint": {
44
  "name": "Engineering strain at maximum bending stress",
45
  "unit": "%",
46
  "definition_en": "Engineering strain at the same sampled point as the largest bending stress, using the earliest occurrence if maxima repeat."
47
+ }
 
48
  }
49
  ],
50
  "scoring_method": "spearman",
 
56
  "spearman_tie_method": "average",
57
  "spearman_undefined_policy": "task_error",
58
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
59
  "format_errors": {
60
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
61
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
62
  },
63
  "max_submission_bytes": 1048576,
64
  "minimum_conditions_per_group": 2,
 
75
  "report_path": "/logs/artifacts/report.md",
76
  "sources_path": "/logs/artifacts/sources.json",
77
  "analysis_path": "/logs/artifacts/analysis/",
78
+ "forbidden_result_scope": "Experimental bending stress-strain, flexural failure, toughness or equivalent mechanical rankings for Ag-Cu-Te-Se-S chalcogenide alloy composition and vacancy series are same-type results and prohibited, including related composition or test-protocol variants. General solid mechanics and independent chemical or crystallographic basic facts outside these result panels are allowed.",
79
+ "supporting_artifact_review": {
80
+ "policy_stop_marker": "SPU_POLICY_STOP",
81
+ "required_files": [
82
+ "report.md",
83
+ "sources.json"
84
+ ],
85
+ "required_directory": "analysis",
86
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
87
+ }
88
  }
tasks/agcute-bending-endpoints-ranking/environment/public/experiment.md CHANGED
@@ -22,7 +22,7 @@ Follow the stated synthesis and three-point bending test. For each original engi
22
 
23
  ## Item settings
24
 
25
- Expand each item in targets.csv or protocol.json together with the common conditions. Every selected item is listed using the same fields.
26
 
27
  ## Source limitations relevant to the definition
28
 
 
22
 
23
  ## Item settings
24
 
25
+ Expand each item in targets.csv or protocol.json together with the common conditions.
26
 
27
  ## Source limitations relevant to the definition
28
 
tasks/agcute-bending-endpoints-ranking/environment/public/information_policy.md CHANGED
@@ -9,3 +9,5 @@ Forbidden examples:
9
  - The source study or its supplementary material; equivalent experimental screening values or rankings; predictors trained or calibrated on those target or same-type results.
10
 
11
  Use audited tools for all searches, downloads and code-initiated network requests. Check both requests and returned content before use. Do not circumvent access controls or ask another model/service for forbidden results. If a permitted query unexpectedly exposes prohibited external results, immediately stop prediction and further search; record only the triggering step, source and stopping reason in report.md without restating results. Do not retrieve or reconstruct specific experimental results from your own model memory; briefly disclose a noticed recollection in your report and continue using permitted inputs and reasoning.
 
 
 
9
  - The source study or its supplementary material; equivalent experimental screening values or rankings; predictors trained or calibrated on those target or same-type results.
10
 
11
  Use audited tools for all searches, downloads and code-initiated network requests. Check both requests and returned content before use. Do not circumvent access controls or ask another model/service for forbidden results. If a permitted query unexpectedly exposes prohibited external results, immediately stop prediction and further search; record only the triggering step, source and stopping reason in report.md without restating results. Do not retrieve or reconstruct specific experimental results from your own model memory; briefly disclose a noticed recollection in your report and continue using permitted inputs and reasoning.
12
+
13
+ For a stop under the information policy, put `SPU_POLICY_STOP` on the first line of `report.md` and record the triggering step, source and reason below it. Withdraw any provisional ranking.
tasks/agcute-bending-endpoints-ranking/environment/public/protocol.json CHANGED
@@ -27,25 +27,34 @@
27
  "measurement_rule": "Use measurements_by_group for each group. The measurement field is a default summary and does not override a group-specific endpoint.",
28
  "inheritance_rule": "common_conditions apply unless the same field is explicitly overridden by the item; item settings are literal experimental facts, not predicted outcomes.",
29
  "items": {
30
- "exp_65960918c5e1": {
31
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
32
  1,
33
  1,
34
- 0.85,
35
  0.1,
36
- 0.05
37
  ]
38
  },
39
- "exp_0b361eedc092": {
40
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
41
  1,
42
  1,
43
- 0.8,
44
  0.1,
45
- 0.1
 
 
 
 
 
 
 
 
 
46
  ]
47
  },
48
- "exp_0a801637748b": {
49
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
50
  1,
51
  1,
@@ -54,16 +63,16 @@
54
  0.12
55
  ]
56
  },
57
- "exp_39fd1eed6b1c": {
58
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
59
  1,
60
  1,
61
- 0.7,
62
  0.1,
63
- 0.2
64
  ]
65
  },
66
- "exp_365507fb236c": {
67
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
68
  0.998,
69
  0.998,
@@ -72,70 +81,79 @@
72
  0.1
73
  ]
74
  },
75
- "exp_3c3c8522da9d": {
76
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
77
  1,
78
  1,
79
- 0.82,
80
- 0.1,
81
- 0.08
82
  ]
83
  },
84
- "exp_fc4984656462": {
85
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
86
  1,
87
  1,
88
- 0.75,
89
  0.1,
90
- 0.15
91
  ]
92
  },
93
- "exp_3fed9f8f07c9": {
94
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
95
  1,
96
  1,
97
- 1,
98
- 0,
99
- 0
100
  ]
101
  },
102
- "exp_aaa9e2c6f5b9": {
103
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
104
  1,
105
  1,
106
- 0.9,
107
  0.1,
108
- 0
109
  ]
110
  },
111
- "exp_54b8fc8b3571": {
112
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
113
  1,
114
  1,
115
- 0.85,
116
  0.1,
117
- 0.05
118
  ]
119
  },
120
- "exp_3aa8fedf978b": {
121
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
122
- 0.998,
123
- 0.998,
124
- 0.8,
125
  0.1,
126
- 0.1
127
  ]
128
  },
129
- "exp_2d9d2ad06f6e": {
130
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
131
  1,
132
  1,
 
 
 
 
 
 
 
 
 
133
  0.8,
134
  0.1,
135
  0.1
136
  ]
137
  },
138
- "exp_8cfb0d14a259": {
139
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
140
  1,
141
  1,
@@ -144,7 +162,7 @@
144
  0.12
145
  ]
146
  },
147
- "exp_d53afffea8ad": {
148
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
149
  1,
150
  1,
@@ -153,25 +171,7 @@
153
  0
154
  ]
155
  },
156
- "exp_69a2a5342ec6": {
157
- "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
158
- 1,
159
- 1,
160
- 1,
161
- 0,
162
- 0
163
- ]
164
- },
165
- "exp_eb05cc2a9dc8": {
166
- "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
167
- 1,
168
- 1,
169
- 0.7,
170
- 0.1,
171
- 0.2
172
- ]
173
- },
174
- "exp_7b741589c10a": {
175
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
176
  1,
177
  1,
@@ -180,13 +180,13 @@
180
  0.08
181
  ]
182
  },
183
- "exp_4369fc61d66d": {
184
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
185
  1,
186
  1,
187
- 0.75,
188
  0.1,
189
- 0.15
190
  ]
191
  }
192
  }
 
27
  "measurement_rule": "Use measurements_by_group for each group. The measurement field is a default summary and does not override a group-specific endpoint.",
28
  "inheritance_rule": "common_conditions apply unless the same field is explicitly overridden by the item; item settings are literal experimental facts, not predicted outcomes.",
29
  "items": {
30
+ "exp_003fe25c3f91": {
31
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
32
  1,
33
  1,
34
+ 0.8,
35
  0.1,
36
+ 0.1
37
  ]
38
  },
39
+ "exp_224a125b3d5a": {
40
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
41
  1,
42
  1,
43
+ 0.75,
44
  0.1,
45
+ 0.15
46
+ ]
47
+ },
48
+ "exp_63b9d2553a78": {
49
+ "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
50
+ 1,
51
+ 1,
52
+ 0.82,
53
+ 0.1,
54
+ 0.08
55
  ]
56
  },
57
+ "exp_672208100095": {
58
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
59
  1,
60
  1,
 
63
  0.12
64
  ]
65
  },
66
+ "exp_7c4f7caec6cf": {
67
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
68
  1,
69
  1,
70
+ 0.9,
71
  0.1,
72
+ 0
73
  ]
74
  },
75
+ "exp_c339123d0b13": {
76
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
77
  0.998,
78
  0.998,
 
81
  0.1
82
  ]
83
  },
84
+ "exp_cf10a9056f26": {
85
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
86
  1,
87
  1,
88
+ 1,
89
+ 0,
90
+ 0
91
  ]
92
  },
93
+ "exp_ea54530b4221": {
94
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
95
  1,
96
  1,
97
+ 0.85,
98
  0.1,
99
+ 0.05
100
  ]
101
  },
102
+ "exp_ff5031900fc6": {
103
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
104
  1,
105
  1,
106
+ 0.7,
107
+ 0.1,
108
+ 0.2
109
  ]
110
  },
111
+ "exp_0a7a7c7f2b75": {
112
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
113
  1,
114
  1,
115
+ 0.8,
116
  0.1,
117
+ 0.1
118
  ]
119
  },
120
+ "exp_235fb0dc6cd6": {
121
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
122
  1,
123
  1,
124
+ 0.75,
125
  0.1,
126
+ 0.15
127
  ]
128
  },
129
+ "exp_351f1bf17bc1": {
130
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
131
+ 1,
132
+ 1,
133
+ 0.85,
134
  0.1,
135
+ 0.05
136
  ]
137
  },
138
+ "exp_3df219ed4754": {
139
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
140
  1,
141
  1,
142
+ 1,
143
+ 0,
144
+ 0
145
+ ]
146
+ },
147
+ "exp_6cc95c89ecb5": {
148
+ "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
149
+ 0.998,
150
+ 0.998,
151
  0.8,
152
  0.1,
153
  0.1
154
  ]
155
  },
156
+ "exp_7ae22fdad3ab": {
157
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
158
  1,
159
  1,
 
162
  0.12
163
  ]
164
  },
165
+ "exp_84f36130f06f": {
166
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
167
  1,
168
  1,
 
171
  0
172
  ]
173
  },
174
+ "exp_d1a81b569dde": {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
175
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
176
  1,
177
  1,
 
180
  0.08
181
  ]
182
  },
183
+ "exp_fb7f19041d5a": {
184
  "nominal_atomic_ratio_Ag_Cu_Te_Se_S": [
185
  1,
186
  1,
187
+ 0.7,
188
  0.1,
189
+ 0.2
190
  ]
191
  }
192
  }
tasks/agcute-bending-endpoints-ranking/environment/public/targets.csv CHANGED
@@ -1,19 +1,19 @@
1
  group_id,item_id,settings_ref,nominal_atomic_ratio_Ag_Cu_Te_Se_S
2
- max_stress,exp_65960918c5e1,protocol.json#exp_65960918c5e1,"[1, 1, 0.85, 0.1, 0.05]"
3
- max_stress,exp_0b361eedc092,protocol.json#exp_0b361eedc092,"[1, 1, 0.8, 0.1, 0.1]"
4
- max_stress,exp_0a801637748b,protocol.json#exp_0a801637748b,"[1, 1, 0.78, 0.1, 0.12]"
5
- max_stress,exp_39fd1eed6b1c,protocol.json#exp_39fd1eed6b1c,"[1, 1, 0.7, 0.1, 0.2]"
6
- max_stress,exp_365507fb236c,protocol.json#exp_365507fb236c,"[0.998, 0.998, 0.8, 0.1, 0.1]"
7
- max_stress,exp_3c3c8522da9d,protocol.json#exp_3c3c8522da9d,"[1, 1, 0.82, 0.1, 0.08]"
8
- max_stress,exp_fc4984656462,protocol.json#exp_fc4984656462,"[1, 1, 0.75, 0.1, 0.15]"
9
- max_stress,exp_3fed9f8f07c9,protocol.json#exp_3fed9f8f07c9,"[1, 1, 1, 0, 0]"
10
- max_stress,exp_aaa9e2c6f5b9,protocol.json#exp_aaa9e2c6f5b9,"[1, 1, 0.9, 0.1, 0]"
11
- strain_at_max,exp_54b8fc8b3571,protocol.json#exp_54b8fc8b3571,"[1, 1, 0.85, 0.1, 0.05]"
12
- strain_at_max,exp_3aa8fedf978b,protocol.json#exp_3aa8fedf978b,"[0.998, 0.998, 0.8, 0.1, 0.1]"
13
- strain_at_max,exp_2d9d2ad06f6e,protocol.json#exp_2d9d2ad06f6e,"[1, 1, 0.8, 0.1, 0.1]"
14
- strain_at_max,exp_8cfb0d14a259,protocol.json#exp_8cfb0d14a259,"[1, 1, 0.78, 0.1, 0.12]"
15
- strain_at_max,exp_d53afffea8ad,protocol.json#exp_d53afffea8ad,"[1, 1, 0.9, 0.1, 0]"
16
- strain_at_max,exp_69a2a5342ec6,protocol.json#exp_69a2a5342ec6,"[1, 1, 1, 0, 0]"
17
- strain_at_max,exp_eb05cc2a9dc8,protocol.json#exp_eb05cc2a9dc8,"[1, 1, 0.7, 0.1, 0.2]"
18
- strain_at_max,exp_7b741589c10a,protocol.json#exp_7b741589c10a,"[1, 1, 0.82, 0.1, 0.08]"
19
- strain_at_max,exp_4369fc61d66d,protocol.json#exp_4369fc61d66d,"[1, 1, 0.75, 0.1, 0.15]"
 
1
  group_id,item_id,settings_ref,nominal_atomic_ratio_Ag_Cu_Te_Se_S
2
+ max_stress,exp_003fe25c3f91,protocol.json#exp_003fe25c3f91,"[1, 1, 0.8, 0.1, 0.1]"
3
+ max_stress,exp_224a125b3d5a,protocol.json#exp_224a125b3d5a,"[1, 1, 0.75, 0.1, 0.15]"
4
+ max_stress,exp_63b9d2553a78,protocol.json#exp_63b9d2553a78,"[1, 1, 0.82, 0.1, 0.08]"
5
+ max_stress,exp_672208100095,protocol.json#exp_672208100095,"[1, 1, 0.78, 0.1, 0.12]"
6
+ max_stress,exp_7c4f7caec6cf,protocol.json#exp_7c4f7caec6cf,"[1, 1, 0.9, 0.1, 0]"
7
+ max_stress,exp_c339123d0b13,protocol.json#exp_c339123d0b13,"[0.998, 0.998, 0.8, 0.1, 0.1]"
8
+ max_stress,exp_cf10a9056f26,protocol.json#exp_cf10a9056f26,"[1, 1, 1, 0, 0]"
9
+ max_stress,exp_ea54530b4221,protocol.json#exp_ea54530b4221,"[1, 1, 0.85, 0.1, 0.05]"
10
+ max_stress,exp_ff5031900fc6,protocol.json#exp_ff5031900fc6,"[1, 1, 0.7, 0.1, 0.2]"
11
+ strain_at_max,exp_0a7a7c7f2b75,protocol.json#exp_0a7a7c7f2b75,"[1, 1, 0.8, 0.1, 0.1]"
12
+ strain_at_max,exp_235fb0dc6cd6,protocol.json#exp_235fb0dc6cd6,"[1, 1, 0.75, 0.1, 0.15]"
13
+ strain_at_max,exp_351f1bf17bc1,protocol.json#exp_351f1bf17bc1,"[1, 1, 0.85, 0.1, 0.05]"
14
+ strain_at_max,exp_3df219ed4754,protocol.json#exp_3df219ed4754,"[1, 1, 1, 0, 0]"
15
+ strain_at_max,exp_6cc95c89ecb5,protocol.json#exp_6cc95c89ecb5,"[0.998, 0.998, 0.8, 0.1, 0.1]"
16
+ strain_at_max,exp_7ae22fdad3ab,protocol.json#exp_7ae22fdad3ab,"[1, 1, 0.78, 0.1, 0.12]"
17
+ strain_at_max,exp_84f36130f06f,protocol.json#exp_84f36130f06f,"[1, 1, 0.9, 0.1, 0]"
18
+ strain_at_max,exp_d1a81b569dde,protocol.json#exp_d1a81b569dde,"[1, 1, 0.82, 0.1, 0.08]"
19
+ strain_at_max,exp_fb7f19041d5a,protocol.json#exp_fb7f19041d5a,"[1, 1, 0.7, 0.1, 0.2]"
tasks/agcute-bending-endpoints-ranking/instruction.md CHANGED
@@ -20,7 +20,7 @@ Rank each comparison group from lowest to highest using that group's specified e
20
 
21
  ## Experimental definition
22
 
23
- Read `/workspace/public/experiment.md`, `protocol.json` and `targets.csv`. The common conditions in protocol.json apply to every item unless that item's setting explicitly overrides the same field. Expand both parts together. Public sequences, structures and other supplied assets are part of the definition. Do not obtain missing details from the prohibited original study.
24
 
25
  Follow the stated synthesis and three-point bending test. For each original engineering stress-strain trace, use its largest reported bending-stress sample and the strain at that same sample; if the maximum is repeated, use the earliest recorded occurrence. Negative pre-load samples are retained in the trace. Trace endpoints are not treated as fracture measurements.
26
 
@@ -52,112 +52,99 @@ You have 8 CPU cores, 16,384 MiB system RAM and one NVIDIA H20 GPU, shared by al
52
 
53
  ## Submission
54
 
55
- Write UTF-8 JSON to `/logs/artifacts/rankings.json` (at most 1 MiB, unique object keys). Include the exact task identity and exactly one entry per group; each order must contain every assigned item once, from lowest to highest. The following is a format example in public display order, not a predicted answer:
56
 
57
  ```json
58
  {
59
  "task_id": "agcute-bending-endpoints-ranking",
60
- "task_version": "1.0.0",
61
  "rankings": [
62
  {
63
  "group_id": "max_stress",
64
  "order": [
65
- "exp_65960918c5e1",
66
- "exp_0b361eedc092",
67
- "exp_0a801637748b",
68
- "exp_39fd1eed6b1c",
69
- "exp_365507fb236c",
70
- "exp_3c3c8522da9d",
71
- "exp_fc4984656462",
72
- "exp_3fed9f8f07c9",
73
- "exp_aaa9e2c6f5b9"
74
  ]
75
  },
76
  {
77
  "group_id": "strain_at_max",
78
  "order": [
79
- "exp_54b8fc8b3571",
80
- "exp_3aa8fedf978b",
81
- "exp_2d9d2ad06f6e",
82
- "exp_8cfb0d14a259",
83
- "exp_d53afffea8ad",
84
- "exp_69a2a5342ec6",
85
- "exp_eb05cc2a9dc8",
86
- "exp_7b741589c10a",
87
- "exp_4369fc61d66d"
88
  ]
89
  }
90
  ]
91
  }
92
  ```
93
 
94
- Write an English account of scientific reasoning, calculations, assumptions and uncertainty to `/logs/artifacts/report.md`; source records to `/logs/artifacts/sources.json`; supporting scripts and calculations to `/logs/artifacts/analysis/`. The evaluator uses the last submission received before the deadline. Empty or missing submissions receive core -1 and auxiliary 0.
95
 
96
- ## Ranking scores
97
 
98
- The primary score is standard Spearman rank correlation, from -1 to 1, computed separately in each group and combined with fixed group weights. Equal measured values receive average true ranks; predictions remain strict permutations. Correct/reverse rankings score 1/-1 when true values have no ties; with true ties the best strict ranking can score less than 1. Random complete rankings have expected correlation 0.
99
 
100
- The auxiliary score counts each decidable pair equally within its group, combines group accuracies using the same weights, then applies one whole-task random 95th-percentile threshold. Known uncertainty uses the declared difference-error rule; unknown uncertainty uses reported values with epsilon and does not exclude unequal results. The threshold is hidden. Core and auxiliary scores are recorded separately; neither meaning of zero proves the absence of predictive ability. Detailed algorithms and all format penalties are in the following authoritative public contract, from which the companion files are generated.
101
-
102
- <!-- public-contract -->
103
  ```json
104
  {
105
  "task_id": "agcute-bending-endpoints-ranking",
106
- "task_version": "1.0.0",
107
  "groups": [
108
  {
109
  "group_id": "max_stress",
110
  "label": "Alloy composition series: bending stress",
111
  "item_ids": [
112
- "exp_65960918c5e1",
113
- "exp_0b361eedc092",
114
- "exp_0a801637748b",
115
- "exp_39fd1eed6b1c",
116
- "exp_365507fb236c",
117
- "exp_3c3c8522da9d",
118
- "exp_fc4984656462",
119
- "exp_3fed9f8f07c9",
120
- "exp_aaa9e2c6f5b9"
121
  ],
122
  "direction": "ascending",
123
  "group_weight": 0.5,
124
- "k": 2,
125
- "epsilon": 0,
126
- "uncertainty_model": "optional_reported_uncertainty",
127
- "difference_standard_error": "sem_sum_upper_bound",
128
  "endpoint": {
129
  "name": "Maximum bending stress",
130
  "unit": "MPa",
131
  "definition_en": "Largest recorded engineering bending stress in each curve."
132
- },
133
- "uncertainty_definition": "Reliable independent uncertainty is not available for this panel. Unequal reported values are compared directly; uncertainty is not assumed to be zero."
134
  },
135
  {
136
  "group_id": "strain_at_max",
137
  "label": "Alloy composition series: peak-stress strain",
138
  "item_ids": [
139
- "exp_54b8fc8b3571",
140
- "exp_3aa8fedf978b",
141
- "exp_2d9d2ad06f6e",
142
- "exp_8cfb0d14a259",
143
- "exp_d53afffea8ad",
144
- "exp_69a2a5342ec6",
145
- "exp_eb05cc2a9dc8",
146
- "exp_7b741589c10a",
147
- "exp_4369fc61d66d"
148
  ],
149
  "direction": "ascending",
150
  "group_weight": 0.5,
151
- "k": 2,
152
- "epsilon": 0,
153
- "uncertainty_model": "optional_reported_uncertainty",
154
- "difference_standard_error": "sem_sum_upper_bound",
155
  "endpoint": {
156
  "name": "Engineering strain at maximum bending stress",
157
  "unit": "%",
158
  "definition_en": "Engineering strain at the same sampled point as the largest bending stress, using the earliest occurrence if maxima repeat."
159
- },
160
- "uncertainty_definition": "Reliable independent uncertainty is not available for this panel. Unequal reported values are compared directly; uncertainty is not assumed to be zero."
161
  }
162
  ],
163
  "scoring_method": "spearman",
@@ -169,18 +156,9 @@ The auxiliary score counts each decidable pair equally within its group, combine
169
  "spearman_tie_method": "average",
170
  "spearman_undefined_policy": "task_error",
171
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
172
- "secondary_scoring_method": "equal_pair_inversion_random95",
173
- "ranking_reward_range": [
174
- 0,
175
- 1
176
- ],
177
- "random_quantile": 0.95,
178
- "unknown_uncertainty_policy": "use_reported_value",
179
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
180
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
181
  "format_errors": {
182
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
183
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
184
  },
185
  "max_submission_bytes": 1048576,
186
  "minimum_conditions_per_group": 2,
@@ -197,7 +175,16 @@ The auxiliary score counts each decidable pair equally within its group, combine
197
  "report_path": "/logs/artifacts/report.md",
198
  "sources_path": "/logs/artifacts/sources.json",
199
  "analysis_path": "/logs/artifacts/analysis/",
200
- "forbidden_result_scope": "Experimental bending stress-strain, flexural failure, toughness or equivalent mechanical rankings for Ag-Cu-Te-Se-S chalcogenide alloy composition and vacancy series are same-type results and prohibited, including related composition or test-protocol variants. General solid mechanics and independent chemical or crystallographic basic facts outside these result panels are allowed."
 
 
 
 
 
 
 
 
 
201
  }
202
  ```
203
  <!-- /public-contract -->
 
20
 
21
  ## Experimental definition
22
 
23
+ Read `/workspace/public/experiment.md`, `protocol.json` and `targets.csv`. The common conditions in protocol.json apply to every item unless that item's setting explicitly overrides the same field. Expand both parts together. Do not obtain missing details from the prohibited original study.
24
 
25
  Follow the stated synthesis and three-point bending test. For each original engineering stress-strain trace, use its largest reported bending-stress sample and the strain at that same sample; if the maximum is repeated, use the earliest recorded occurrence. Negative pre-load samples are retained in the trace. Trace endpoints are not treated as fracture measurements.
26
 
 
52
 
53
  ## Submission
54
 
55
+ Write UTF-8 JSON to `/logs/artifacts/rankings.json` (at most 1 MiB, unique object keys). Include the exact task identity and exactly one entry per group; each order must contain every assigned item once, from lowest to highest. The example illustrates the submission format. Submit your predicted order for each group.
56
 
57
  ```json
58
  {
59
  "task_id": "agcute-bending-endpoints-ranking",
60
+ "task_version": "1.1.0",
61
  "rankings": [
62
  {
63
  "group_id": "max_stress",
64
  "order": [
65
+ "exp_003fe25c3f91",
66
+ "exp_224a125b3d5a",
67
+ "exp_63b9d2553a78",
68
+ "exp_672208100095",
69
+ "exp_7c4f7caec6cf",
70
+ "exp_c339123d0b13",
71
+ "exp_cf10a9056f26",
72
+ "exp_ea54530b4221",
73
+ "exp_ff5031900fc6"
74
  ]
75
  },
76
  {
77
  "group_id": "strain_at_max",
78
  "order": [
79
+ "exp_0a7a7c7f2b75",
80
+ "exp_235fb0dc6cd6",
81
+ "exp_351f1bf17bc1",
82
+ "exp_3df219ed4754",
83
+ "exp_6cc95c89ecb5",
84
+ "exp_7ae22fdad3ab",
85
+ "exp_84f36130f06f",
86
+ "exp_d1a81b569dde",
87
+ "exp_fb7f19041d5a"
88
  ]
89
  }
90
  ]
91
  }
92
  ```
93
 
94
+ Write an English account of scientific reasoning, calculations, assumptions and uncertainty to `/logs/artifacts/report.md`; source records to `/logs/artifacts/sources.json`; supporting scripts and calculations to `/logs/artifacts/analysis/`. Only rankings.json contributes to the Spearman reward. The verifier records the presence of the supporting artifacts for review. For a stop under the information policy, put `SPU_POLICY_STOP` on the first line of `report.md` and record the triggering step, source and reason below it. Withdraw any provisional ranking. The evaluator uses the last submission received before the deadline. Empty or missing submissions receive -1.
95
 
96
+ ## Ranking score
97
 
98
+ The task score is the weighted mean of the group Spearman rank correlations and ranges from -1 to 1. Equal measurements share their average rank. Submit a strict ordering for each group. When measurements are distinct, the measured order scores 1 and its reverse scores -1. Tied measurements can place the highest attainable correlation below 1. A uniformly random ordering has expected correlation 0.
99
 
 
 
 
100
  ```json
101
  {
102
  "task_id": "agcute-bending-endpoints-ranking",
103
+ "task_version": "1.1.0",
104
  "groups": [
105
  {
106
  "group_id": "max_stress",
107
  "label": "Alloy composition series: bending stress",
108
  "item_ids": [
109
+ "exp_003fe25c3f91",
110
+ "exp_224a125b3d5a",
111
+ "exp_63b9d2553a78",
112
+ "exp_672208100095",
113
+ "exp_7c4f7caec6cf",
114
+ "exp_c339123d0b13",
115
+ "exp_cf10a9056f26",
116
+ "exp_ea54530b4221",
117
+ "exp_ff5031900fc6"
118
  ],
119
  "direction": "ascending",
120
  "group_weight": 0.5,
 
 
 
 
121
  "endpoint": {
122
  "name": "Maximum bending stress",
123
  "unit": "MPa",
124
  "definition_en": "Largest recorded engineering bending stress in each curve."
125
+ }
 
126
  },
127
  {
128
  "group_id": "strain_at_max",
129
  "label": "Alloy composition series: peak-stress strain",
130
  "item_ids": [
131
+ "exp_0a7a7c7f2b75",
132
+ "exp_235fb0dc6cd6",
133
+ "exp_351f1bf17bc1",
134
+ "exp_3df219ed4754",
135
+ "exp_6cc95c89ecb5",
136
+ "exp_7ae22fdad3ab",
137
+ "exp_84f36130f06f",
138
+ "exp_d1a81b569dde",
139
+ "exp_fb7f19041d5a"
140
  ],
141
  "direction": "ascending",
142
  "group_weight": 0.5,
 
 
 
 
143
  "endpoint": {
144
  "name": "Engineering strain at maximum bending stress",
145
  "unit": "%",
146
  "definition_en": "Engineering strain at the same sampled point as the largest bending stress, using the earliest occurrence if maxima repeat."
147
+ }
 
148
  }
149
  ],
150
  "scoring_method": "spearman",
 
156
  "spearman_tie_method": "average",
157
  "spearman_undefined_policy": "task_error",
158
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
159
  "format_errors": {
160
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
161
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
162
  },
163
  "max_submission_bytes": 1048576,
164
  "minimum_conditions_per_group": 2,
 
175
  "report_path": "/logs/artifacts/report.md",
176
  "sources_path": "/logs/artifacts/sources.json",
177
  "analysis_path": "/logs/artifacts/analysis/",
178
+ "forbidden_result_scope": "Experimental bending stress-strain, flexural failure, toughness or equivalent mechanical rankings for Ag-Cu-Te-Se-S chalcogenide alloy composition and vacancy series are same-type results and prohibited, including related composition or test-protocol variants. General solid mechanics and independent chemical or crystallographic basic facts outside these result panels are allowed.",
179
+ "supporting_artifact_review": {
180
+ "policy_stop_marker": "SPU_POLICY_STOP",
181
+ "required_files": [
182
+ "report.md",
183
+ "sources.json"
184
+ ],
185
+ "required_directory": "analysis",
186
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
187
+ }
188
  }
189
  ```
190
  <!-- /public-contract -->
tasks/agcute-bending-endpoints-ranking/solution/solve.sh CHANGED
@@ -1,39 +1,39 @@
1
  #!/bin/sh
2
  set -eu
3
  mkdir -p /logs/artifacts
4
- cat > /logs/artifacts/rankings.json <<'PROJECTAA_ORACLE'
5
  {
6
  "task_id": "agcute-bending-endpoints-ranking",
7
- "task_version": "1.0.0",
8
  "rankings": [
9
  {
10
  "group_id": "max_stress",
11
  "order": [
12
- "exp_3fed9f8f07c9",
13
- "exp_aaa9e2c6f5b9",
14
- "exp_3c3c8522da9d",
15
- "exp_fc4984656462",
16
- "exp_0b361eedc092",
17
- "exp_0a801637748b",
18
- "exp_65960918c5e1",
19
- "exp_39fd1eed6b1c",
20
- "exp_365507fb236c"
21
  ]
22
  },
23
  {
24
  "group_id": "strain_at_max",
25
  "order": [
26
- "exp_69a2a5342ec6",
27
- "exp_d53afffea8ad",
28
- "exp_eb05cc2a9dc8",
29
- "exp_4369fc61d66d",
30
- "exp_54b8fc8b3571",
31
- "exp_2d9d2ad06f6e",
32
- "exp_8cfb0d14a259",
33
- "exp_3aa8fedf978b",
34
- "exp_7b741589c10a"
35
  ]
36
  }
37
  ]
38
  }
39
- PROJECTAA_ORACLE
 
1
  #!/bin/sh
2
  set -eu
3
  mkdir -p /logs/artifacts
4
+ cat > /logs/artifacts/rankings.json <<'EOF'
5
  {
6
  "task_id": "agcute-bending-endpoints-ranking",
7
+ "task_version": "1.1.0",
8
  "rankings": [
9
  {
10
  "group_id": "max_stress",
11
  "order": [
12
+ "exp_cf10a9056f26",
13
+ "exp_7c4f7caec6cf",
14
+ "exp_63b9d2553a78",
15
+ "exp_224a125b3d5a",
16
+ "exp_003fe25c3f91",
17
+ "exp_672208100095",
18
+ "exp_ea54530b4221",
19
+ "exp_ff5031900fc6",
20
+ "exp_c339123d0b13"
21
  ]
22
  },
23
  {
24
  "group_id": "strain_at_max",
25
  "order": [
26
+ "exp_3df219ed4754",
27
+ "exp_84f36130f06f",
28
+ "exp_fb7f19041d5a",
29
+ "exp_235fb0dc6cd6",
30
+ "exp_351f1bf17bc1",
31
+ "exp_0a7a7c7f2b75",
32
+ "exp_7ae22fdad3ab",
33
+ "exp_6cc95c89ecb5",
34
+ "exp_d1a81b569dde"
35
  ]
36
  }
37
  ]
38
  }
39
+ EOF
tasks/agcute-bending-endpoints-ranking/task.toml CHANGED
@@ -1,9 +1,7 @@
1
- schema_version = "1.2"
2
  [metadata]
3
  task_id = "agcute-bending-endpoints-ranking"
4
- task_version = "1.0.0"
5
- pipeline_version = "19"
6
- scientist_prompt_version = "projectaa-scientist-en-v7-ranking"
7
  [agent]
8
  user = "solver"
9
  timeout_sec = 28800.0
@@ -14,7 +12,7 @@ memory_mb = 16384
14
  storage_mb = 20480
15
  gpus = 1
16
  gpu_types = ["H20"]
17
- allow_internet = true
18
  [verifier]
19
  timeout_sec = 120.0
20
  environment_mode = "separate"
@@ -24,4 +22,4 @@ cpus = 1
24
  memory_mb = 1024
25
  storage_mb = 2048
26
  gpus = 0
27
- allow_internet = false
 
1
+ schema_version = "1.4"
2
  [metadata]
3
  task_id = "agcute-bending-endpoints-ranking"
4
+ task_version = "1.1.0"
 
 
5
  [agent]
6
  user = "solver"
7
  timeout_sec = 28800.0
 
12
  storage_mb = 20480
13
  gpus = 1
14
  gpu_types = ["H20"]
15
+ network_mode = "public"
16
  [verifier]
17
  timeout_sec = 120.0
18
  environment_mode = "separate"
 
22
  memory_mb = 1024
23
  storage_mb = 2048
24
  gpus = 0
25
+ network_mode = "no-network"
tasks/agcute-bending-endpoints-ranking/tests/answers.json CHANGED
@@ -1,924 +1,142 @@
1
  {
2
  "task_id": "agcute-bending-endpoints-ranking",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "max_stress",
7
  "items": [
8
  {
9
- "item_id": "exp_65960918c5e1",
10
- "mean": 225.883,
11
- "n": null,
12
- "uncertainty_status": "unknown",
13
- "uncertainty_scale": null,
14
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
15
- "quality_status": "retained_with_flag",
16
- "warnings": [
17
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
18
- ],
19
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 70960, strain column 7, stress column 8; maximum over 81959 paired curve points",
20
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
21
- },
22
- {
23
- "item_id": "exp_0b361eedc092",
24
  "mean": 203.947,
25
  "n": null,
26
  "uncertainty_status": "unknown",
27
- "uncertainty_scale": null,
28
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
29
- "quality_status": "retained_with_flag",
30
- "warnings": [
31
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
32
- ],
33
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 112839, strain column 13, stress column 14; maximum over 186708 paired curve points",
34
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
35
  },
36
  {
37
- "item_id": "exp_0a801637748b",
38
- "mean": 208.152,
39
  "n": null,
40
  "uncertainty_status": "unknown",
41
- "uncertainty_scale": null,
42
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
43
- "quality_status": "retained_with_flag",
44
- "warnings": [
45
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
46
- ],
47
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 119513, strain column 16, stress column 17; maximum over 141872 paired curve points",
48
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
49
  },
50
  {
51
- "item_id": "exp_39fd1eed6b1c",
52
- "mean": 234.551,
53
  "n": null,
54
  "uncertainty_status": "unknown",
55
- "uncertainty_scale": null,
56
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
57
- "quality_status": "retained_with_flag",
58
- "warnings": [
59
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
60
- ],
61
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 7608, strain column 22, stress column 23; maximum over 7607 paired curve points",
62
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
63
  },
64
  {
65
- "item_id": "exp_365507fb236c",
66
- "mean": 255.958,
67
  "n": null,
68
  "uncertainty_status": "unknown",
69
- "uncertainty_scale": null,
70
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
71
- "quality_status": "retained_with_flag",
72
- "warnings": [
73
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
74
- ],
75
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 118267, strain column 25, stress column 26; maximum over 156878 paired curve points",
76
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
77
  },
78
  {
79
- "item_id": "exp_3c3c8522da9d",
80
- "mean": 184.302,
81
  "n": null,
82
  "uncertainty_status": "unknown",
83
- "uncertainty_scale": null,
84
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
85
- "quality_status": "retained_with_flag",
86
- "warnings": [
87
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
88
- ],
89
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 116064, strain column 10, stress column 11; maximum over 126540 paired curve points",
90
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
91
  },
92
  {
93
- "item_id": "exp_fc4984656462",
94
- "mean": 197.459,
95
  "n": null,
96
  "uncertainty_status": "unknown",
97
- "uncertainty_scale": null,
98
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
99
- "quality_status": "retained_with_flag",
100
- "warnings": [
101
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
102
- ],
103
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 64453, strain column 19, stress column 20; maximum over 64459 paired curve points",
104
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
105
  },
106
  {
107
- "item_id": "exp_3fed9f8f07c9",
108
  "mean": 61.4139,
109
  "n": null,
110
  "uncertainty_status": "unknown",
111
- "uncertainty_scale": null,
112
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
113
- "quality_status": "retained_with_flag",
114
- "warnings": [
115
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
116
- ],
117
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 5162, strain column 1, stress column 2; maximum over 5162 paired curve points",
118
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
119
  },
120
  {
121
- "item_id": "exp_aaa9e2c6f5b9",
122
- "mean": 148.722,
123
  "n": null,
124
  "uncertainty_status": "unknown",
125
- "uncertainty_scale": null,
126
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
127
- "quality_status": "retained_with_flag",
128
- "warnings": [
129
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
130
- ],
131
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 13882, strain column 4, stress column 5; maximum over 13881 paired curve points",
132
- "aggregation": "Maximum of all finite paired stress samples in the original curve; no interpolation."
133
- }
134
- ],
135
- "pairs": [
136
- {
137
- "item_i": "exp_65960918c5e1",
138
- "item_j": "exp_0b361eedc092",
139
- "mean_difference": 21.936000000000007,
140
- "difference_se": null,
141
- "uncertainty_status": "unknown",
142
- "comparison_basis": "reported_value",
143
- "comparable": 1
144
- },
145
- {
146
- "item_i": "exp_65960918c5e1",
147
- "item_j": "exp_0a801637748b",
148
- "mean_difference": 17.731000000000023,
149
- "difference_se": null,
150
- "uncertainty_status": "unknown",
151
- "comparison_basis": "reported_value",
152
- "comparable": 1
153
- },
154
- {
155
- "item_i": "exp_65960918c5e1",
156
- "item_j": "exp_39fd1eed6b1c",
157
- "mean_difference": -8.667999999999978,
158
- "difference_se": null,
159
- "uncertainty_status": "unknown",
160
- "comparison_basis": "reported_value",
161
- "comparable": 1
162
- },
163
- {
164
- "item_i": "exp_65960918c5e1",
165
- "item_j": "exp_365507fb236c",
166
- "mean_difference": -30.07499999999999,
167
- "difference_se": null,
168
- "uncertainty_status": "unknown",
169
- "comparison_basis": "reported_value",
170
- "comparable": 1
171
- },
172
- {
173
- "item_i": "exp_65960918c5e1",
174
- "item_j": "exp_3c3c8522da9d",
175
- "mean_difference": 41.58100000000002,
176
- "difference_se": null,
177
- "uncertainty_status": "unknown",
178
- "comparison_basis": "reported_value",
179
- "comparable": 1
180
- },
181
- {
182
- "item_i": "exp_65960918c5e1",
183
- "item_j": "exp_fc4984656462",
184
- "mean_difference": 28.424000000000007,
185
- "difference_se": null,
186
- "uncertainty_status": "unknown",
187
- "comparison_basis": "reported_value",
188
- "comparable": 1
189
- },
190
- {
191
- "item_i": "exp_65960918c5e1",
192
- "item_j": "exp_3fed9f8f07c9",
193
- "mean_difference": 164.46910000000003,
194
- "difference_se": null,
195
- "uncertainty_status": "unknown",
196
- "comparison_basis": "reported_value",
197
- "comparable": 1
198
- },
199
- {
200
- "item_i": "exp_65960918c5e1",
201
- "item_j": "exp_aaa9e2c6f5b9",
202
- "mean_difference": 77.161,
203
- "difference_se": null,
204
- "uncertainty_status": "unknown",
205
- "comparison_basis": "reported_value",
206
- "comparable": 1
207
- },
208
- {
209
- "item_i": "exp_0b361eedc092",
210
- "item_j": "exp_0a801637748b",
211
- "mean_difference": -4.204999999999984,
212
- "difference_se": null,
213
- "uncertainty_status": "unknown",
214
- "comparison_basis": "reported_value",
215
- "comparable": 1
216
- },
217
- {
218
- "item_i": "exp_0b361eedc092",
219
- "item_j": "exp_39fd1eed6b1c",
220
- "mean_difference": -30.603999999999985,
221
- "difference_se": null,
222
- "uncertainty_status": "unknown",
223
- "comparison_basis": "reported_value",
224
- "comparable": 1
225
- },
226
- {
227
- "item_i": "exp_0b361eedc092",
228
- "item_j": "exp_365507fb236c",
229
- "mean_difference": -52.010999999999996,
230
- "difference_se": null,
231
- "uncertainty_status": "unknown",
232
- "comparison_basis": "reported_value",
233
- "comparable": 1
234
- },
235
- {
236
- "item_i": "exp_0b361eedc092",
237
- "item_j": "exp_3c3c8522da9d",
238
- "mean_difference": 19.64500000000001,
239
- "difference_se": null,
240
- "uncertainty_status": "unknown",
241
- "comparison_basis": "reported_value",
242
- "comparable": 1
243
- },
244
- {
245
- "item_i": "exp_0b361eedc092",
246
- "item_j": "exp_fc4984656462",
247
- "mean_difference": 6.4879999999999995,
248
- "difference_se": null,
249
- "uncertainty_status": "unknown",
250
- "comparison_basis": "reported_value",
251
- "comparable": 1
252
- },
253
- {
254
- "item_i": "exp_0b361eedc092",
255
- "item_j": "exp_3fed9f8f07c9",
256
- "mean_difference": 142.5331,
257
- "difference_se": null,
258
- "uncertainty_status": "unknown",
259
- "comparison_basis": "reported_value",
260
- "comparable": 1
261
- },
262
- {
263
- "item_i": "exp_0b361eedc092",
264
- "item_j": "exp_aaa9e2c6f5b9",
265
- "mean_difference": 55.224999999999994,
266
- "difference_se": null,
267
- "uncertainty_status": "unknown",
268
- "comparison_basis": "reported_value",
269
- "comparable": 1
270
  },
271
  {
272
- "item_i": "exp_0a801637748b",
273
- "item_j": "exp_39fd1eed6b1c",
274
- "mean_difference": -26.399,
275
- "difference_se": null,
276
- "uncertainty_status": "unknown",
277
- "comparison_basis": "reported_value",
278
- "comparable": 1
279
- },
280
- {
281
- "item_i": "exp_0a801637748b",
282
- "item_j": "exp_365507fb236c",
283
- "mean_difference": -47.80600000000001,
284
- "difference_se": null,
285
- "uncertainty_status": "unknown",
286
- "comparison_basis": "reported_value",
287
- "comparable": 1
288
- },
289
- {
290
- "item_i": "exp_0a801637748b",
291
- "item_j": "exp_3c3c8522da9d",
292
- "mean_difference": 23.849999999999994,
293
- "difference_se": null,
294
- "uncertainty_status": "unknown",
295
- "comparison_basis": "reported_value",
296
- "comparable": 1
297
- },
298
- {
299
- "item_i": "exp_0a801637748b",
300
- "item_j": "exp_fc4984656462",
301
- "mean_difference": 10.692999999999984,
302
- "difference_se": null,
303
- "uncertainty_status": "unknown",
304
- "comparison_basis": "reported_value",
305
- "comparable": 1
306
- },
307
- {
308
- "item_i": "exp_0a801637748b",
309
- "item_j": "exp_3fed9f8f07c9",
310
- "mean_difference": 146.73809999999997,
311
- "difference_se": null,
312
- "uncertainty_status": "unknown",
313
- "comparison_basis": "reported_value",
314
- "comparable": 1
315
- },
316
- {
317
- "item_i": "exp_0a801637748b",
318
- "item_j": "exp_aaa9e2c6f5b9",
319
- "mean_difference": 59.42999999999998,
320
- "difference_se": null,
321
- "uncertainty_status": "unknown",
322
- "comparison_basis": "reported_value",
323
- "comparable": 1
324
- },
325
- {
326
- "item_i": "exp_39fd1eed6b1c",
327
- "item_j": "exp_365507fb236c",
328
- "mean_difference": -21.40700000000001,
329
- "difference_se": null,
330
- "uncertainty_status": "unknown",
331
- "comparison_basis": "reported_value",
332
- "comparable": 1
333
- },
334
- {
335
- "item_i": "exp_39fd1eed6b1c",
336
- "item_j": "exp_3c3c8522da9d",
337
- "mean_difference": 50.248999999999995,
338
- "difference_se": null,
339
- "uncertainty_status": "unknown",
340
- "comparison_basis": "reported_value",
341
- "comparable": 1
342
- },
343
- {
344
- "item_i": "exp_39fd1eed6b1c",
345
- "item_j": "exp_fc4984656462",
346
- "mean_difference": 37.091999999999985,
347
- "difference_se": null,
348
- "uncertainty_status": "unknown",
349
- "comparison_basis": "reported_value",
350
- "comparable": 1
351
- },
352
- {
353
- "item_i": "exp_39fd1eed6b1c",
354
- "item_j": "exp_3fed9f8f07c9",
355
- "mean_difference": 173.13709999999998,
356
- "difference_se": null,
357
- "uncertainty_status": "unknown",
358
- "comparison_basis": "reported_value",
359
- "comparable": 1
360
- },
361
- {
362
- "item_i": "exp_39fd1eed6b1c",
363
- "item_j": "exp_aaa9e2c6f5b9",
364
- "mean_difference": 85.82899999999998,
365
- "difference_se": null,
366
- "uncertainty_status": "unknown",
367
- "comparison_basis": "reported_value",
368
- "comparable": 1
369
- },
370
- {
371
- "item_i": "exp_365507fb236c",
372
- "item_j": "exp_3c3c8522da9d",
373
- "mean_difference": 71.656,
374
- "difference_se": null,
375
- "uncertainty_status": "unknown",
376
- "comparison_basis": "reported_value",
377
- "comparable": 1
378
- },
379
- {
380
- "item_i": "exp_365507fb236c",
381
- "item_j": "exp_fc4984656462",
382
- "mean_difference": 58.498999999999995,
383
- "difference_se": null,
384
- "uncertainty_status": "unknown",
385
- "comparison_basis": "reported_value",
386
- "comparable": 1
387
- },
388
- {
389
- "item_i": "exp_365507fb236c",
390
- "item_j": "exp_3fed9f8f07c9",
391
- "mean_difference": 194.54410000000001,
392
- "difference_se": null,
393
- "uncertainty_status": "unknown",
394
- "comparison_basis": "reported_value",
395
- "comparable": 1
396
- },
397
- {
398
- "item_i": "exp_365507fb236c",
399
- "item_j": "exp_aaa9e2c6f5b9",
400
- "mean_difference": 107.23599999999999,
401
- "difference_se": null,
402
- "uncertainty_status": "unknown",
403
- "comparison_basis": "reported_value",
404
- "comparable": 1
405
- },
406
- {
407
- "item_i": "exp_3c3c8522da9d",
408
- "item_j": "exp_fc4984656462",
409
- "mean_difference": -13.15700000000001,
410
- "difference_se": null,
411
- "uncertainty_status": "unknown",
412
- "comparison_basis": "reported_value",
413
- "comparable": 1
414
- },
415
- {
416
- "item_i": "exp_3c3c8522da9d",
417
- "item_j": "exp_3fed9f8f07c9",
418
- "mean_difference": 122.8881,
419
- "difference_se": null,
420
- "uncertainty_status": "unknown",
421
- "comparison_basis": "reported_value",
422
- "comparable": 1
423
- },
424
- {
425
- "item_i": "exp_3c3c8522da9d",
426
- "item_j": "exp_aaa9e2c6f5b9",
427
- "mean_difference": 35.579999999999984,
428
- "difference_se": null,
429
- "uncertainty_status": "unknown",
430
- "comparison_basis": "reported_value",
431
- "comparable": 1
432
- },
433
- {
434
- "item_i": "exp_fc4984656462",
435
- "item_j": "exp_3fed9f8f07c9",
436
- "mean_difference": 136.0451,
437
- "difference_se": null,
438
- "uncertainty_status": "unknown",
439
- "comparison_basis": "reported_value",
440
- "comparable": 1
441
- },
442
- {
443
- "item_i": "exp_fc4984656462",
444
- "item_j": "exp_aaa9e2c6f5b9",
445
- "mean_difference": 48.736999999999995,
446
- "difference_se": null,
447
- "uncertainty_status": "unknown",
448
- "comparison_basis": "reported_value",
449
- "comparable": 1
450
- },
451
- {
452
- "item_i": "exp_3fed9f8f07c9",
453
- "item_j": "exp_aaa9e2c6f5b9",
454
- "mean_difference": -87.30810000000001,
455
- "difference_se": null,
456
  "uncertainty_status": "unknown",
457
- "comparison_basis": "reported_value",
458
- "comparable": 1
459
  }
460
- ],
461
- "comparable_pairs": 36,
462
- "quality_status": "passed"
463
  },
464
  {
465
  "group_id": "strain_at_max",
466
  "items": [
467
  {
468
- "item_id": "exp_54b8fc8b3571",
469
- "mean": 3.58553,
470
  "n": null,
471
  "uncertainty_status": "unknown",
472
- "uncertainty_scale": null,
473
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
474
- "quality_status": "retained_with_flag",
475
- "warnings": [
476
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
477
- ],
478
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 70960, strain column 7, stress column 8; maximum over 81959 paired curve points",
479
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
480
  },
481
  {
482
- "item_id": "exp_3aa8fedf978b",
483
- "mean": 6.20628,
484
  "n": null,
485
  "uncertainty_status": "unknown",
486
- "uncertainty_scale": null,
487
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
488
- "quality_status": "retained_with_flag",
489
- "warnings": [
490
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
491
- ],
492
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 118267, strain column 25, stress column 26; maximum over 156878 paired curve points",
493
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
494
  },
495
  {
496
- "item_id": "exp_2d9d2ad06f6e",
497
- "mean": 5.59208,
498
  "n": null,
499
  "uncertainty_status": "unknown",
500
- "uncertainty_scale": null,
501
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
502
- "quality_status": "retained_with_flag",
503
- "warnings": [
504
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
505
- ],
506
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 112839, strain column 13, stress column 14; maximum over 186708 paired curve points",
507
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
508
  },
509
  {
510
- "item_id": "exp_8cfb0d14a259",
511
- "mean": 6.15589,
512
  "n": null,
513
  "uncertainty_status": "unknown",
514
- "uncertainty_scale": null,
515
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
516
- "quality_status": "retained_with_flag",
517
- "warnings": [
518
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
519
- ],
520
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 119513, strain column 16, stress column 17; maximum over 141872 paired curve points",
521
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
522
  },
523
  {
524
- "item_id": "exp_d53afffea8ad",
525
- "mean": 0.67426,
526
  "n": null,
527
  "uncertainty_status": "unknown",
528
- "uncertainty_scale": null,
529
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
530
- "quality_status": "retained_with_flag",
531
- "warnings": [
532
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
533
- ],
534
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 13882, strain column 4, stress column 5; maximum over 13881 paired curve points",
535
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
536
  },
537
  {
538
- "item_id": "exp_69a2a5342ec6",
539
- "mean": 0.26561,
540
  "n": null,
541
  "uncertainty_status": "unknown",
542
- "uncertainty_scale": null,
543
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
544
- "quality_status": "retained_with_flag",
545
- "warnings": [
546
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
547
- ],
548
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 5162, strain column 1, stress column 2; maximum over 5162 paired curve points",
549
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
550
  },
551
  {
552
- "item_id": "exp_eb05cc2a9dc8",
553
- "mean": 1.57783,
554
  "n": null,
555
  "uncertainty_status": "unknown",
556
- "uncertainty_scale": null,
557
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
558
- "quality_status": "retained_with_flag",
559
- "warnings": [
560
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
561
- ],
562
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 7608, strain column 22, stress column 23; maximum over 7607 paired curve points",
563
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
564
  },
565
  {
566
- "item_id": "exp_7b741589c10a",
567
  "mean": 6.31575,
568
  "n": null,
569
  "uncertainty_status": "unknown",
570
- "uncertainty_scale": null,
571
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
572
- "quality_status": "retained_with_flag",
573
- "warnings": [
574
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
575
- ],
576
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 116064, strain column 10, stress column 11; maximum over 126540 paired curve points",
577
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
578
  },
579
  {
580
- "item_id": "exp_4369fc61d66d",
581
- "mean": 3.19442,
582
  "n": null,
583
  "uncertainty_status": "unknown",
584
- "uncertainty_scale": null,
585
- "uncertainty_basis": "No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.",
586
- "quality_status": "retained_with_flag",
587
- "warnings": [
588
- "Unknown independent uncertainty; use reported values without inventing zero uncertainty."
589
- ],
590
- "source_location": "Original-SourceData.xlsx 'Figure 1b' row 64453, strain column 19, stress column 20; maximum over 64459 paired curve points",
591
- "aggregation": "Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation."
592
- }
593
- ],
594
- "pairs": [
595
- {
596
- "item_i": "exp_54b8fc8b3571",
597
- "item_j": "exp_3aa8fedf978b",
598
- "mean_difference": -2.6207499999999997,
599
- "difference_se": null,
600
- "uncertainty_status": "unknown",
601
- "comparison_basis": "reported_value",
602
- "comparable": 1
603
- },
604
- {
605
- "item_i": "exp_54b8fc8b3571",
606
- "item_j": "exp_2d9d2ad06f6e",
607
- "mean_difference": -2.0065500000000003,
608
- "difference_se": null,
609
- "uncertainty_status": "unknown",
610
- "comparison_basis": "reported_value",
611
- "comparable": 1
612
- },
613
- {
614
- "item_i": "exp_54b8fc8b3571",
615
- "item_j": "exp_8cfb0d14a259",
616
- "mean_difference": -2.5703600000000004,
617
- "difference_se": null,
618
- "uncertainty_status": "unknown",
619
- "comparison_basis": "reported_value",
620
- "comparable": 1
621
- },
622
- {
623
- "item_i": "exp_54b8fc8b3571",
624
- "item_j": "exp_d53afffea8ad",
625
- "mean_difference": 2.91127,
626
- "difference_se": null,
627
- "uncertainty_status": "unknown",
628
- "comparison_basis": "reported_value",
629
- "comparable": 1
630
- },
631
- {
632
- "item_i": "exp_54b8fc8b3571",
633
- "item_j": "exp_69a2a5342ec6",
634
- "mean_difference": 3.3199199999999998,
635
- "difference_se": null,
636
- "uncertainty_status": "unknown",
637
- "comparison_basis": "reported_value",
638
- "comparable": 1
639
- },
640
- {
641
- "item_i": "exp_54b8fc8b3571",
642
- "item_j": "exp_eb05cc2a9dc8",
643
- "mean_difference": 2.0077,
644
- "difference_se": null,
645
- "uncertainty_status": "unknown",
646
- "comparison_basis": "reported_value",
647
- "comparable": 1
648
- },
649
- {
650
- "item_i": "exp_54b8fc8b3571",
651
- "item_j": "exp_7b741589c10a",
652
- "mean_difference": -2.7302200000000005,
653
- "difference_se": null,
654
- "uncertainty_status": "unknown",
655
- "comparison_basis": "reported_value",
656
- "comparable": 1
657
- },
658
- {
659
- "item_i": "exp_54b8fc8b3571",
660
- "item_j": "exp_4369fc61d66d",
661
- "mean_difference": 0.39110999999999985,
662
- "difference_se": null,
663
- "uncertainty_status": "unknown",
664
- "comparison_basis": "reported_value",
665
- "comparable": 1
666
- },
667
- {
668
- "item_i": "exp_3aa8fedf978b",
669
- "item_j": "exp_2d9d2ad06f6e",
670
- "mean_difference": 0.6141999999999994,
671
- "difference_se": null,
672
- "uncertainty_status": "unknown",
673
- "comparison_basis": "reported_value",
674
- "comparable": 1
675
- },
676
- {
677
- "item_i": "exp_3aa8fedf978b",
678
- "item_j": "exp_8cfb0d14a259",
679
- "mean_difference": 0.05038999999999927,
680
- "difference_se": null,
681
- "uncertainty_status": "unknown",
682
- "comparison_basis": "reported_value",
683
- "comparable": 1
684
- },
685
- {
686
- "item_i": "exp_3aa8fedf978b",
687
- "item_j": "exp_d53afffea8ad",
688
- "mean_difference": 5.532019999999999,
689
- "difference_se": null,
690
- "uncertainty_status": "unknown",
691
- "comparison_basis": "reported_value",
692
- "comparable": 1
693
- },
694
- {
695
- "item_i": "exp_3aa8fedf978b",
696
- "item_j": "exp_69a2a5342ec6",
697
- "mean_difference": 5.94067,
698
- "difference_se": null,
699
- "uncertainty_status": "unknown",
700
- "comparison_basis": "reported_value",
701
- "comparable": 1
702
- },
703
- {
704
- "item_i": "exp_3aa8fedf978b",
705
- "item_j": "exp_eb05cc2a9dc8",
706
- "mean_difference": 4.628449999999999,
707
- "difference_se": null,
708
- "uncertainty_status": "unknown",
709
- "comparison_basis": "reported_value",
710
- "comparable": 1
711
- },
712
- {
713
- "item_i": "exp_3aa8fedf978b",
714
- "item_j": "exp_7b741589c10a",
715
- "mean_difference": -0.10947000000000084,
716
- "difference_se": null,
717
- "uncertainty_status": "unknown",
718
- "comparison_basis": "reported_value",
719
- "comparable": 1
720
- },
721
- {
722
- "item_i": "exp_3aa8fedf978b",
723
- "item_j": "exp_4369fc61d66d",
724
- "mean_difference": 3.0118599999999995,
725
- "difference_se": null,
726
- "uncertainty_status": "unknown",
727
- "comparison_basis": "reported_value",
728
- "comparable": 1
729
- },
730
- {
731
- "item_i": "exp_2d9d2ad06f6e",
732
- "item_j": "exp_8cfb0d14a259",
733
- "mean_difference": -0.5638100000000001,
734
- "difference_se": null,
735
- "uncertainty_status": "unknown",
736
- "comparison_basis": "reported_value",
737
- "comparable": 1
738
- },
739
- {
740
- "item_i": "exp_2d9d2ad06f6e",
741
- "item_j": "exp_d53afffea8ad",
742
- "mean_difference": 4.91782,
743
- "difference_se": null,
744
- "uncertainty_status": "unknown",
745
- "comparison_basis": "reported_value",
746
- "comparable": 1
747
- },
748
- {
749
- "item_i": "exp_2d9d2ad06f6e",
750
- "item_j": "exp_69a2a5342ec6",
751
- "mean_difference": 5.3264700000000005,
752
- "difference_se": null,
753
- "uncertainty_status": "unknown",
754
- "comparison_basis": "reported_value",
755
- "comparable": 1
756
- },
757
- {
758
- "item_i": "exp_2d9d2ad06f6e",
759
- "item_j": "exp_eb05cc2a9dc8",
760
- "mean_difference": 4.0142500000000005,
761
- "difference_se": null,
762
- "uncertainty_status": "unknown",
763
- "comparison_basis": "reported_value",
764
- "comparable": 1
765
- },
766
- {
767
- "item_i": "exp_2d9d2ad06f6e",
768
- "item_j": "exp_7b741589c10a",
769
- "mean_difference": -0.7236700000000003,
770
- "difference_se": null,
771
- "uncertainty_status": "unknown",
772
- "comparison_basis": "reported_value",
773
- "comparable": 1
774
- },
775
- {
776
- "item_i": "exp_2d9d2ad06f6e",
777
- "item_j": "exp_4369fc61d66d",
778
- "mean_difference": 2.39766,
779
- "difference_se": null,
780
- "uncertainty_status": "unknown",
781
- "comparison_basis": "reported_value",
782
- "comparable": 1
783
- },
784
- {
785
- "item_i": "exp_8cfb0d14a259",
786
- "item_j": "exp_d53afffea8ad",
787
- "mean_difference": 5.48163,
788
- "difference_se": null,
789
- "uncertainty_status": "unknown",
790
- "comparison_basis": "reported_value",
791
- "comparable": 1
792
- },
793
- {
794
- "item_i": "exp_8cfb0d14a259",
795
- "item_j": "exp_69a2a5342ec6",
796
- "mean_difference": 5.890280000000001,
797
- "difference_se": null,
798
- "uncertainty_status": "unknown",
799
- "comparison_basis": "reported_value",
800
- "comparable": 1
801
- },
802
- {
803
- "item_i": "exp_8cfb0d14a259",
804
- "item_j": "exp_eb05cc2a9dc8",
805
- "mean_difference": 4.578060000000001,
806
- "difference_se": null,
807
- "uncertainty_status": "unknown",
808
- "comparison_basis": "reported_value",
809
- "comparable": 1
810
- },
811
- {
812
- "item_i": "exp_8cfb0d14a259",
813
- "item_j": "exp_7b741589c10a",
814
- "mean_difference": -0.1598600000000001,
815
- "difference_se": null,
816
- "uncertainty_status": "unknown",
817
- "comparison_basis": "reported_value",
818
- "comparable": 1
819
- },
820
- {
821
- "item_i": "exp_8cfb0d14a259",
822
- "item_j": "exp_4369fc61d66d",
823
- "mean_difference": 2.9614700000000003,
824
- "difference_se": null,
825
- "uncertainty_status": "unknown",
826
- "comparison_basis": "reported_value",
827
- "comparable": 1
828
- },
829
- {
830
- "item_i": "exp_d53afffea8ad",
831
- "item_j": "exp_69a2a5342ec6",
832
- "mean_difference": 0.40864999999999996,
833
- "difference_se": null,
834
- "uncertainty_status": "unknown",
835
- "comparison_basis": "reported_value",
836
- "comparable": 1
837
- },
838
- {
839
- "item_i": "exp_d53afffea8ad",
840
- "item_j": "exp_eb05cc2a9dc8",
841
- "mean_difference": -0.9035700000000001,
842
- "difference_se": null,
843
- "uncertainty_status": "unknown",
844
- "comparison_basis": "reported_value",
845
- "comparable": 1
846
- },
847
- {
848
- "item_i": "exp_d53afffea8ad",
849
- "item_j": "exp_7b741589c10a",
850
- "mean_difference": -5.64149,
851
- "difference_se": null,
852
- "uncertainty_status": "unknown",
853
- "comparison_basis": "reported_value",
854
- "comparable": 1
855
- },
856
- {
857
- "item_i": "exp_d53afffea8ad",
858
- "item_j": "exp_4369fc61d66d",
859
- "mean_difference": -2.52016,
860
- "difference_se": null,
861
- "uncertainty_status": "unknown",
862
- "comparison_basis": "reported_value",
863
- "comparable": 1
864
- },
865
- {
866
- "item_i": "exp_69a2a5342ec6",
867
- "item_j": "exp_eb05cc2a9dc8",
868
- "mean_difference": -1.31222,
869
- "difference_se": null,
870
- "uncertainty_status": "unknown",
871
- "comparison_basis": "reported_value",
872
- "comparable": 1
873
- },
874
- {
875
- "item_i": "exp_69a2a5342ec6",
876
- "item_j": "exp_7b741589c10a",
877
- "mean_difference": -6.050140000000001,
878
- "difference_se": null,
879
- "uncertainty_status": "unknown",
880
- "comparison_basis": "reported_value",
881
- "comparable": 1
882
- },
883
- {
884
- "item_i": "exp_69a2a5342ec6",
885
- "item_j": "exp_4369fc61d66d",
886
- "mean_difference": -2.92881,
887
- "difference_se": null,
888
- "uncertainty_status": "unknown",
889
- "comparison_basis": "reported_value",
890
- "comparable": 1
891
- },
892
- {
893
- "item_i": "exp_eb05cc2a9dc8",
894
- "item_j": "exp_7b741589c10a",
895
- "mean_difference": -4.737920000000001,
896
- "difference_se": null,
897
- "uncertainty_status": "unknown",
898
- "comparison_basis": "reported_value",
899
- "comparable": 1
900
- },
901
- {
902
- "item_i": "exp_eb05cc2a9dc8",
903
- "item_j": "exp_4369fc61d66d",
904
- "mean_difference": -1.61659,
905
- "difference_se": null,
906
- "uncertainty_status": "unknown",
907
- "comparison_basis": "reported_value",
908
- "comparable": 1
909
- },
910
- {
911
- "item_i": "exp_7b741589c10a",
912
- "item_j": "exp_4369fc61d66d",
913
- "mean_difference": 3.1213300000000004,
914
- "difference_se": null,
915
- "uncertainty_status": "unknown",
916
- "comparison_basis": "reported_value",
917
- "comparable": 1
918
  }
919
- ],
920
- "comparable_pairs": 36,
921
- "quality_status": "passed"
922
  }
923
  ]
924
  }
 
1
  {
2
  "task_id": "agcute-bending-endpoints-ranking",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "max_stress",
7
  "items": [
8
  {
9
+ "item_id": "exp_003fe25c3f91",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
10
  "mean": 203.947,
11
  "n": null,
12
  "uncertainty_status": "unknown",
13
+ "uncertainty_scale": null
 
 
 
 
 
 
 
14
  },
15
  {
16
+ "item_id": "exp_224a125b3d5a",
17
+ "mean": 197.459,
18
  "n": null,
19
  "uncertainty_status": "unknown",
20
+ "uncertainty_scale": null
 
 
 
 
 
 
 
21
  },
22
  {
23
+ "item_id": "exp_63b9d2553a78",
24
+ "mean": 184.302,
25
  "n": null,
26
  "uncertainty_status": "unknown",
27
+ "uncertainty_scale": null
 
 
 
 
 
 
 
28
  },
29
  {
30
+ "item_id": "exp_672208100095",
31
+ "mean": 208.152,
32
  "n": null,
33
  "uncertainty_status": "unknown",
34
+ "uncertainty_scale": null
 
 
 
 
 
 
 
35
  },
36
  {
37
+ "item_id": "exp_7c4f7caec6cf",
38
+ "mean": 148.722,
39
  "n": null,
40
  "uncertainty_status": "unknown",
41
+ "uncertainty_scale": null
 
 
 
 
 
 
 
42
  },
43
  {
44
+ "item_id": "exp_c339123d0b13",
45
+ "mean": 255.958,
46
  "n": null,
47
  "uncertainty_status": "unknown",
48
+ "uncertainty_scale": null
 
 
 
 
 
 
 
49
  },
50
  {
51
+ "item_id": "exp_cf10a9056f26",
52
  "mean": 61.4139,
53
  "n": null,
54
  "uncertainty_status": "unknown",
55
+ "uncertainty_scale": null
 
 
 
 
 
 
 
56
  },
57
  {
58
+ "item_id": "exp_ea54530b4221",
59
+ "mean": 225.883,
60
  "n": null,
61
  "uncertainty_status": "unknown",
62
+ "uncertainty_scale": null
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
63
  },
64
  {
65
+ "item_id": "exp_ff5031900fc6",
66
+ "mean": 234.551,
67
+ "n": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
68
  "uncertainty_status": "unknown",
69
+ "uncertainty_scale": null
 
70
  }
71
+ ]
 
 
72
  },
73
  {
74
  "group_id": "strain_at_max",
75
  "items": [
76
  {
77
+ "item_id": "exp_0a7a7c7f2b75",
78
+ "mean": 5.59208,
79
  "n": null,
80
  "uncertainty_status": "unknown",
81
+ "uncertainty_scale": null
 
 
 
 
 
 
 
82
  },
83
  {
84
+ "item_id": "exp_235fb0dc6cd6",
85
+ "mean": 3.19442,
86
  "n": null,
87
  "uncertainty_status": "unknown",
88
+ "uncertainty_scale": null
 
 
 
 
 
 
 
89
  },
90
  {
91
+ "item_id": "exp_351f1bf17bc1",
92
+ "mean": 3.58553,
93
  "n": null,
94
  "uncertainty_status": "unknown",
95
+ "uncertainty_scale": null
 
 
 
 
 
 
 
96
  },
97
  {
98
+ "item_id": "exp_3df219ed4754",
99
+ "mean": 0.26561,
100
  "n": null,
101
  "uncertainty_status": "unknown",
102
+ "uncertainty_scale": null
 
 
 
 
 
 
 
103
  },
104
  {
105
+ "item_id": "exp_6cc95c89ecb5",
106
+ "mean": 6.20628,
107
  "n": null,
108
  "uncertainty_status": "unknown",
109
+ "uncertainty_scale": null
 
 
 
 
 
 
 
110
  },
111
  {
112
+ "item_id": "exp_7ae22fdad3ab",
113
+ "mean": 6.15589,
114
  "n": null,
115
  "uncertainty_status": "unknown",
116
+ "uncertainty_scale": null
 
 
 
 
 
 
 
117
  },
118
  {
119
+ "item_id": "exp_84f36130f06f",
120
+ "mean": 0.67426,
121
  "n": null,
122
  "uncertainty_status": "unknown",
123
+ "uncertainty_scale": null
 
 
 
 
 
 
 
124
  },
125
  {
126
+ "item_id": "exp_d1a81b569dde",
127
  "mean": 6.31575,
128
  "n": null,
129
  "uncertainty_status": "unknown",
130
+ "uncertainty_scale": null
 
 
 
 
 
 
 
131
  },
132
  {
133
+ "item_id": "exp_fb7f19041d5a",
134
+ "mean": 1.57783,
135
  "n": null,
136
  "uncertainty_status": "unknown",
137
+ "uncertainty_scale": null
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
138
  }
139
+ ]
 
 
140
  }
141
  ]
142
  }
tasks/agcute-bending-endpoints-ranking/tests/contract.json CHANGED
@@ -1,60 +1,50 @@
1
  {
2
  "task_id": "agcute-bending-endpoints-ranking",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "max_stress",
7
  "label": "Alloy composition series: bending stress",
8
  "item_ids": [
9
- "exp_65960918c5e1",
10
- "exp_0b361eedc092",
11
- "exp_0a801637748b",
12
- "exp_39fd1eed6b1c",
13
- "exp_365507fb236c",
14
- "exp_3c3c8522da9d",
15
- "exp_fc4984656462",
16
- "exp_3fed9f8f07c9",
17
- "exp_aaa9e2c6f5b9"
18
  ],
19
  "direction": "ascending",
20
  "group_weight": 0.5,
21
- "k": 2,
22
- "epsilon": 0,
23
- "uncertainty_model": "optional_reported_uncertainty",
24
- "difference_standard_error": "sem_sum_upper_bound",
25
  "endpoint": {
26
  "name": "Maximum bending stress",
27
  "unit": "MPa",
28
  "definition_en": "Largest recorded engineering bending stress in each curve."
29
- },
30
- "uncertainty_definition": "Reliable independent uncertainty is not available for this panel. Unequal reported values are compared directly; uncertainty is not assumed to be zero."
31
  },
32
  {
33
  "group_id": "strain_at_max",
34
  "label": "Alloy composition series: peak-stress strain",
35
  "item_ids": [
36
- "exp_54b8fc8b3571",
37
- "exp_3aa8fedf978b",
38
- "exp_2d9d2ad06f6e",
39
- "exp_8cfb0d14a259",
40
- "exp_d53afffea8ad",
41
- "exp_69a2a5342ec6",
42
- "exp_eb05cc2a9dc8",
43
- "exp_7b741589c10a",
44
- "exp_4369fc61d66d"
45
  ],
46
  "direction": "ascending",
47
  "group_weight": 0.5,
48
- "k": 2,
49
- "epsilon": 0,
50
- "uncertainty_model": "optional_reported_uncertainty",
51
- "difference_standard_error": "sem_sum_upper_bound",
52
  "endpoint": {
53
  "name": "Engineering strain at maximum bending stress",
54
  "unit": "%",
55
  "definition_en": "Engineering strain at the same sampled point as the largest bending stress, using the earliest occurrence if maxima repeat."
56
- },
57
- "uncertainty_definition": "Reliable independent uncertainty is not available for this panel. Unequal reported values are compared directly; uncertainty is not assumed to be zero."
58
  }
59
  ],
60
  "scoring_method": "spearman",
@@ -66,18 +56,9 @@
66
  "spearman_tie_method": "average",
67
  "spearman_undefined_policy": "task_error",
68
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
69
- "secondary_scoring_method": "equal_pair_inversion_random95",
70
- "ranking_reward_range": [
71
- 0,
72
- 1
73
- ],
74
- "random_quantile": 0.95,
75
- "unknown_uncertainty_policy": "use_reported_value",
76
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
77
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
78
  "format_errors": {
79
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
80
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
81
  },
82
  "max_submission_bytes": 1048576,
83
  "minimum_conditions_per_group": 2,
@@ -94,5 +75,14 @@
94
  "report_path": "/logs/artifacts/report.md",
95
  "sources_path": "/logs/artifacts/sources.json",
96
  "analysis_path": "/logs/artifacts/analysis/",
97
- "forbidden_result_scope": "Experimental bending stress-strain, flexural failure, toughness or equivalent mechanical rankings for Ag-Cu-Te-Se-S chalcogenide alloy composition and vacancy series are same-type results and prohibited, including related composition or test-protocol variants. General solid mechanics and independent chemical or crystallographic basic facts outside these result panels are allowed."
 
 
 
 
 
 
 
 
 
98
  }
 
1
  {
2
  "task_id": "agcute-bending-endpoints-ranking",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "max_stress",
7
  "label": "Alloy composition series: bending stress",
8
  "item_ids": [
9
+ "exp_003fe25c3f91",
10
+ "exp_224a125b3d5a",
11
+ "exp_63b9d2553a78",
12
+ "exp_672208100095",
13
+ "exp_7c4f7caec6cf",
14
+ "exp_c339123d0b13",
15
+ "exp_cf10a9056f26",
16
+ "exp_ea54530b4221",
17
+ "exp_ff5031900fc6"
18
  ],
19
  "direction": "ascending",
20
  "group_weight": 0.5,
 
 
 
 
21
  "endpoint": {
22
  "name": "Maximum bending stress",
23
  "unit": "MPa",
24
  "definition_en": "Largest recorded engineering bending stress in each curve."
25
+ }
 
26
  },
27
  {
28
  "group_id": "strain_at_max",
29
  "label": "Alloy composition series: peak-stress strain",
30
  "item_ids": [
31
+ "exp_0a7a7c7f2b75",
32
+ "exp_235fb0dc6cd6",
33
+ "exp_351f1bf17bc1",
34
+ "exp_3df219ed4754",
35
+ "exp_6cc95c89ecb5",
36
+ "exp_7ae22fdad3ab",
37
+ "exp_84f36130f06f",
38
+ "exp_d1a81b569dde",
39
+ "exp_fb7f19041d5a"
40
  ],
41
  "direction": "ascending",
42
  "group_weight": 0.5,
 
 
 
 
43
  "endpoint": {
44
  "name": "Engineering strain at maximum bending stress",
45
  "unit": "%",
46
  "definition_en": "Engineering strain at the same sampled point as the largest bending stress, using the earliest occurrence if maxima repeat."
47
+ }
 
48
  }
49
  ],
50
  "scoring_method": "spearman",
 
56
  "spearman_tie_method": "average",
57
  "spearman_undefined_policy": "task_error",
58
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
59
  "format_errors": {
60
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
61
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
62
  },
63
  "max_submission_bytes": 1048576,
64
  "minimum_conditions_per_group": 2,
 
75
  "report_path": "/logs/artifacts/report.md",
76
  "sources_path": "/logs/artifacts/sources.json",
77
  "analysis_path": "/logs/artifacts/analysis/",
78
+ "forbidden_result_scope": "Experimental bending stress-strain, flexural failure, toughness or equivalent mechanical rankings for Ag-Cu-Te-Se-S chalcogenide alloy composition and vacancy series are same-type results and prohibited, including related composition or test-protocol variants. General solid mechanics and independent chemical or crystallographic basic facts outside these result panels are allowed.",
79
+ "supporting_artifact_review": {
80
+ "policy_stop_marker": "SPU_POLICY_STOP",
81
+ "required_files": [
82
+ "report.md",
83
+ "sources.json"
84
+ ],
85
+ "required_directory": "analysis",
86
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
87
+ }
88
  }
tasks/agcute-bending-endpoints-ranking/tests/grade.py CHANGED
@@ -1,14 +1,44 @@
1
- """Pipeline v18 Spearman core and preserved equal-pair auxiliary grader."""
2
  import argparse
3
- import copy
4
  import hashlib
5
  import json
6
  import math
 
 
7
  from pathlib import Path
8
 
9
- import original_grade as old
10
 
11
- TaskError = old.TaskError
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
 
13
 
14
  def bounded(value):
@@ -51,98 +81,174 @@ def spearman(values, predicted_positions):
51
  return bounded(math.fsum(r * p for r, p in zip(x, y)) / denominator)
52
 
53
 
54
- def promote_contract(contract):
55
- value = copy.deepcopy(contract)
56
- value.update(scoring_method='spearman', primary_score='task_spearman',
57
- spearman_range=[-1, 1], spearman_tie_method='average',
58
- spearman_undefined_policy='task_error',
59
- secondary_scoring_method='equal_pair_inversion_random95',
60
- ranking_reward_range=[0, 1])
61
- return value
62
-
63
-
64
- def auxiliary_view(contract):
65
- expected = {'scoring_method': 'spearman', 'primary_score': 'task_spearman',
66
- 'spearman_range': [-1, 1], 'spearman_tie_method': 'average',
67
- 'spearman_undefined_policy': 'task_error',
68
- 'secondary_scoring_method': 'equal_pair_inversion_random95',
69
- 'ranking_reward_range': [0, 1]}
70
- if any(contract.get(k) != v for k, v in expected.items()):
71
- raise TaskError('Invalid dual-score contract')
72
- value = copy.deepcopy(contract)
73
- value['scoring_method'] = value.pop('secondary_scoring_method')
74
- for key in ('primary_score', 'spearman_range', 'spearman_tie_method',
75
- 'spearman_undefined_policy', 'ranking_reward_range'):
76
- value.pop(key)
77
- return value
78
-
79
-
80
- def scoring_bindings(contract, answers):
81
- return {'contract_sha256': old.canonical_sha(contract),
82
- 'answers_sha256': old.canonical_sha(answers),
83
- 'grader_sha256': hashlib.sha256(Path(__file__).read_bytes()).hexdigest(),
84
- 'auxiliary_grader_sha256': hashlib.sha256(Path(old.__file__).read_bytes()).hexdigest()}
85
-
86
-
87
- def validate(contract, answers, baseline):
88
- auxiliary = auxiliary_view(contract)
89
- old.validate_task(auxiliary, answers)
90
- if baseline.get('bindings') != scoring_bindings(contract, answers):
91
- raise TaskError('Dual-score baseline hash mismatch')
92
- old_baseline = {'q': baseline.get('q'), 'bindings': old.scoring_bindings(auxiliary, answers)}
93
- old.validate_baseline(auxiliary, answers, old_baseline)
94
- hidden = {g['group_id']: {i['item_id']: i['mean'] for i in g['items']} for g in answers['groups']}
95
- for g in contract['groups']:
96
- spearman([hidden[g['group_id']][i] for i in g['item_ids']], list(range(len(g['item_ids']))))
97
- return auxiliary, old_baseline, hidden
98
-
99
-
100
- def add_core(contract, report, hidden):
101
- report['scoring_method'] = 'spearman'
102
- report['primary_score'] = 'task_spearman'
103
- report['secondary_scoring_method'] = 'equal_pair_inversion_random95'
104
- report['official_reward'] = None
105
- report['official_ranking_reward'] = None
106
- if not report['groups']:
107
- report['task_spearman'] = -1.0
108
- report['core_penalty_reason'] = 'whole_task_format_error'
109
- report['groups'] = [{'group_id': g['group_id'], 'group_weight': g['group_weight'],
110
- 'group_spearman_g': None, 'spearman_for_aggregation': -1.0,
111
- 'group_pair_accuracy': 0.0, 'status': 'whole_task_format_error'}
112
- for g in contract['groups']]
113
- return report
114
- for g in report['groups']:
115
- if g['status'] != 'valid':
116
- rho = None
117
- else:
118
- item_ids = next(c['item_ids'] for c in contract['groups'] if c['group_id'] == g['group_id'])
119
- positions = {item_id: i for i, item_id in enumerate(g['order'])}
120
- rho = spearman([hidden[g['group_id']][i] for i in item_ids], [positions[i] for i in item_ids])
121
- g['group_spearman_g'] = rho
122
- g['spearman_for_aggregation'] = -1.0 if rho is None else rho
123
- report['task_spearman'] = bounded(math.fsum(g['group_weight'] * g['spearman_for_aggregation']
124
- for g in report['groups']))
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
125
  return report
126
 
127
 
128
- def grade_submission(contract, answers, submission, baseline):
129
- auxiliary, old_baseline, hidden = validate(contract, answers, baseline)
130
- report = old.grade_submission(auxiliary, answers, submission, old_baseline)
131
- report['random_q95'] = baseline['q']
132
- return add_core(contract, report, hidden)
133
 
134
 
135
- def grade_file(contract, answers, path, baseline):
136
- auxiliary, old_baseline, hidden = validate(contract, answers, baseline)
137
- return add_core(contract, old.grade_file(auxiliary, answers, path, old_baseline), hidden)
138
 
 
 
 
 
 
 
 
 
 
 
 
 
139
 
140
- def official_scores(report, compliance, execution, memory=False):
141
- if compliance == 'cheating':
142
- return -1.0, 0.0
143
- if compliance != 'compliant' or execution != 'completed':
144
- return None, None
145
- return report['task_spearman'], report['ranking_reward']
 
 
 
 
 
 
 
146
 
147
 
148
  def main():
@@ -156,18 +262,22 @@ def main():
156
  for name in ('reward.txt', 'reward.json', 'report.json', 'evaluator_error.json'):
157
  (args.output / name).unlink(missing_ok=True)
158
  try:
159
- contract, answers, baseline = [old.load_json((args.root / f'{name}.json').read_bytes())
160
- for name in ('contract', 'answers', 'random_baseline')]
161
- report = grade_file(contract, answers, args.submission, baseline)
162
- except (TaskError, old.SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
163
- old.dump(args.output / 'evaluator_error.json', {'status': 'task_error', 'error': str(error)})
 
 
 
 
164
  raise SystemExit(2) from error
165
- old.dump(args.output / 'report.json', report)
 
166
  if args.interface == 'txt':
167
  (args.output / 'reward.txt').write_text(str(report['task_spearman']) + '\n')
168
  else:
169
- old.dump(args.output / 'reward.json', {'reward': report['task_spearman'],
170
- 'ranking_reward': report['ranking_reward']})
171
 
172
 
173
  if __name__ == '__main__':
 
1
+ """Score complete experimental rankings by weighted Spearman correlation."""
2
  import argparse
3
+ import errno
4
  import hashlib
5
  import json
6
  import math
7
+ import os
8
+ import stat
9
  from pathlib import Path
10
 
 
11
 
12
+ class TaskError(ValueError):
13
+ pass
14
+
15
+
16
+ class SubmissionError(ValueError):
17
+ pass
18
+
19
+
20
+ def unique_object(pairs):
21
+ result = {}
22
+ for key, value in pairs:
23
+ if key in result:
24
+ raise SubmissionError("duplicate_json_key")
25
+ result[key] = value
26
+ return result
27
+
28
+
29
+ def load_json(raw):
30
+ def reject_constant(value):
31
+ raise SubmissionError("nonfinite_json_constant")
32
+ return json.loads(raw.decode("utf-8"), object_pairs_hook=unique_object,
33
+ parse_constant=reject_constant)
34
+
35
+
36
+ def finite(value, name, minimum=None):
37
+ if isinstance(value, bool) or not isinstance(value, (int, float)):
38
+ raise TaskError(name + " is not numeric")
39
+ if not math.isfinite(value) or (minimum is not None and value < minimum):
40
+ raise TaskError(name + " is out of range")
41
+ return float(value)
42
 
43
 
44
  def bounded(value):
 
81
  return bounded(math.fsum(r * p for r, p in zip(x, y)) / denominator)
82
 
83
 
84
+ def read_regular(path, limit):
85
+ try:
86
+ fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
87
+ try:
88
+ info = os.fstat(fd)
89
+ if not stat.S_ISREG(info.st_mode):
90
+ raise SubmissionError("not_regular_file")
91
+ if info.st_size > limit:
92
+ raise SubmissionError("submission_too_large")
93
+ with os.fdopen(fd, "rb", closefd=False) as stream:
94
+ raw = stream.read(limit + 1)
95
+ if len(raw) > limit:
96
+ raise SubmissionError("submission_too_large")
97
+ return raw
98
+ finally:
99
+ os.close(fd)
100
+ except OSError as error:
101
+ if error.errno in (errno.ENOENT, errno.ELOOP, errno.EISDIR, errno.ENOTDIR,
102
+ errno.ENXIO, errno.EACCES, errno.EPERM):
103
+ raise SubmissionError("missing_or_unreadable_submission") from error
104
+ raise
105
+
106
+
107
+ def validate_task(contract, answers):
108
+ expected = {
109
+ "scoring_method": "spearman",
110
+ "primary_score": "task_spearman",
111
+ "spearman_range": [-1, 1],
112
+ "spearman_tie_method": "average",
113
+ "spearman_undefined_policy": "task_error",
114
+ }
115
+ if any(contract.get(key) != value for key, value in expected.items()):
116
+ raise TaskError("Unsupported scoring contract")
117
+ for key in ("task_id", "task_version"):
118
+ if not isinstance(contract.get(key), str) or answers.get(key) != contract[key]:
119
+ raise TaskError("Invalid task identity")
120
+ limit = contract.get("max_submission_bytes")
121
+ if type(limit) is not int or not 0 < limit <= 1048576:
122
+ raise TaskError("Invalid submission size limit")
123
+ minimum = contract.get("minimum_conditions_per_group", 2)
124
+ if type(minimum) is not int or minimum < 2:
125
+ raise TaskError("Invalid minimum group size")
126
+ groups = contract["groups"]
127
+ references = {g["group_id"]: g for g in answers["groups"]}
128
+ if not groups or len(groups) != len(references) or len(references) != len(answers["groups"]):
129
+ raise TaskError("Invalid reference groups")
130
+
131
+ measurements = {}
132
+ all_ids = set()
133
+ for group in groups:
134
+ gid = group["group_id"]
135
+ ids = group["item_ids"]
136
+ if not isinstance(gid, str) or not gid or gid in measurements or gid not in references:
137
+ raise TaskError("Invalid group ID")
138
+ if (group["direction"] != "ascending" or not isinstance(ids, list) or len(ids) < minimum
139
+ or any(not isinstance(i, str) or not i for i in ids)
140
+ or len(set(ids)) != len(ids) or all_ids.intersection(ids)):
141
+ raise TaskError("Invalid group definition")
142
+ all_ids.update(ids)
143
+ if finite(group["group_weight"], "group_weight", 0) == 0:
144
+ raise TaskError("Zero group weight")
145
+ items = references[gid]["items"]
146
+ values = {item["item_id"]: item["mean"] for item in items}
147
+ if len(items) != len(ids) or set(values) != set(ids):
148
+ raise TaskError("Reference items differ from the contract")
149
+ measurements[gid] = [values[i] for i in ids]
150
+ spearman(measurements[gid], list(range(len(ids))))
151
+ if not math.isclose(math.fsum(g["group_weight"] for g in groups), 1.0, rel_tol=0, abs_tol=1e-12):
152
+ raise TaskError("Group weights must sum to one")
153
+ return measurements
154
+
155
+
156
+ def group_entries(submission, contract):
157
+ if not isinstance(submission, dict):
158
+ raise SubmissionError("invalid_top_level_structure")
159
+ if any(submission.get(key) != contract[key] for key in ("task_id", "task_version")):
160
+ raise SubmissionError("task_identity_mismatch")
161
+ if not isinstance(submission.get("rankings"), list):
162
+ raise SubmissionError("rankings_not_array")
163
+ entries = {group["group_id"]: [] for group in contract["groups"]}
164
+ for entry in submission["rankings"]:
165
+ if not isinstance(entry, dict) or not isinstance(entry.get("group_id"), str):
166
+ raise SubmissionError("unassignable_group_entry")
167
+ if entry["group_id"] not in entries:
168
+ raise SubmissionError("unknown_group_id")
169
+ entries[entry["group_id"]].append(entry)
170
+ return entries
171
+
172
+
173
+ def score_group(group, entries, values):
174
+ gid = group["group_id"]
175
+ ids = group["item_ids"]
176
+ result = {"group_id": gid, "group_weight": group["group_weight"],
177
+ "group_spearman_g": None, "spearman_for_aggregation": -1.0}
178
+ if entries is None:
179
+ result["status"] = "whole_task_format_error"
180
+ return result
181
+ matches = entries[gid]
182
+ if len(matches) != 1:
183
+ result["status"] = "duplicate_group" if matches else "missing_group"
184
+ return result
185
+ order = matches[0].get("order")
186
+ if (not isinstance(order, list) or len(order) != len(ids)
187
+ or any(not isinstance(i, str) for i in order) or set(order) != set(ids)):
188
+ result["status"] = "invalid_permutation"
189
+ return result
190
+ positions = {item: p for p, item in enumerate(order)}
191
+ rho = spearman(values, [positions[i] for i in ids])
192
+ result.update(status="valid", order=order, group_spearman_g=rho, spearman_for_aggregation=rho)
193
+ return result
194
+
195
+
196
+ def grade_file(contract, answers, path):
197
+ measurements = validate_task(contract, answers)
198
+ report = {"task_id": contract["task_id"], "task_version": contract["task_version"],
199
+ "scoring_method": "spearman", "primary_score": "task_spearman",
200
+ "evaluation_status": "completed"}
201
+ entries = None
202
+ try:
203
+ raw = read_regular(path, contract["max_submission_bytes"])
204
+ submission = load_json(raw)
205
+ report["submission_sha256"] = hashlib.sha256(raw).hexdigest()
206
+ entries = group_entries(submission, contract)
207
+ except (SubmissionError, UnicodeError, json.JSONDecodeError, RecursionError) as error:
208
+ report["submission_status"] = str(error) or type(error).__name__
209
+ report["core_penalty_reason"] = "whole_task_format_error"
210
+ groups = [score_group(g, entries, measurements[g["group_id"]]) for g in contract["groups"]]
211
+ report["groups"] = groups
212
+ if entries is None:
213
+ report["task_spearman"] = -1.0
214
+ else:
215
+ report["submission_status"] = "valid" if all(g["status"] == "valid" for g in groups) else "group_errors"
216
+ report["task_spearman"] = bounded(math.fsum(g["group_weight"] * g["spearman_for_aggregation"]
217
+ for g in groups))
218
  return report
219
 
220
 
221
+ def dump(path, value):
222
+ path.write_text(json.dumps(value, indent=2, allow_nan=False) + "\n", encoding="utf-8")
 
 
 
223
 
224
 
 
 
 
225
 
226
+ def supporting_artifact_status(submission):
227
+ root = submission.parent
228
+ report_path = root / 'report.md'
229
+ paths = {'report_file': report_path, 'sources_file': root / 'sources.json',
230
+ 'analysis_directory': root / 'analysis'}
231
+ present = {}
232
+ for name, path in paths.items():
233
+ try:
234
+ mode = path.lstat().st_mode
235
+ present[name] = stat.S_ISDIR(mode) if name == 'analysis_directory' else stat.S_ISREG(mode)
236
+ except OSError:
237
+ present[name] = False
238
 
239
+ declared_stop = False
240
+ if present['report_file']:
241
+ try:
242
+ fd = os.open(report_path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
243
+ with os.fdopen(fd, 'rb') as stream:
244
+ if stat.S_ISREG(os.fstat(stream.fileno()).st_mode):
245
+ declared_stop = stream.readline(256).rstrip(b'\r\n') == b'SPU_POLICY_STOP'
246
+ except OSError:
247
+ pass
248
+ status = 'artifacts_present' if all(present.values()) else 'artifacts_incomplete'
249
+ if declared_stop:
250
+ status = 'declared_stop_with_ranking' if os.path.lexists(submission) else 'declared_stop'
251
+ return {'status': status, **present, 'policy_stop_declared': declared_stop}
252
 
253
 
254
  def main():
 
262
  for name in ('reward.txt', 'reward.json', 'report.json', 'evaluator_error.json'):
263
  (args.output / name).unlink(missing_ok=True)
264
  try:
265
+ bindings = load_json((args.root / 'scoring_hashes.json').read_bytes())
266
+ for name in ('contract.json', 'answers.json', 'grade.py'):
267
+ if hashlib.sha256((args.root / name).read_bytes()).hexdigest() != bindings[name]:
268
+ raise TaskError('Scoring file hash mismatch: ' + name)
269
+ contract = load_json((args.root / 'contract.json').read_bytes())
270
+ answers = load_json((args.root / 'answers.json').read_bytes())
271
+ report = grade_file(contract, answers, args.submission)
272
+ except (TaskError, SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
273
+ dump(args.output / 'evaluator_error.json', {'status': 'task_error', 'error': str(error)})
274
  raise SystemExit(2) from error
275
+ report['supporting_artifact_review'] = supporting_artifact_status(args.submission)
276
+ dump(args.output / 'report.json', report)
277
  if args.interface == 'txt':
278
  (args.output / 'reward.txt').write_text(str(report['task_spearman']) + '\n')
279
  else:
280
+ dump(args.output / 'reward.json', {'reward': report['task_spearman']})
 
281
 
282
 
283
  if __name__ == '__main__':
tasks/agcute-bending-endpoints-ranking/tests/measurements.csv CHANGED
@@ -1,19 +1,19 @@
1
  group_id,item_id,mean,n,uncertainty_status,uncertainty_scale,uncertainty_basis,quality_status,warnings,source_location,aggregation
2
- max_stress,exp_65960918c5e1,225.883,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 70960, strain column 7, stress column 8; maximum over 81959 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
3
- max_stress,exp_0b361eedc092,203.947,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 112839, strain column 13, stress column 14; maximum over 186708 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
4
- max_stress,exp_0a801637748b,208.152,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 119513, strain column 16, stress column 17; maximum over 141872 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
5
- max_stress,exp_39fd1eed6b1c,234.551,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 7608, strain column 22, stress column 23; maximum over 7607 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
6
- max_stress,exp_365507fb236c,255.958,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 118267, strain column 25, stress column 26; maximum over 156878 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
7
- max_stress,exp_3c3c8522da9d,184.302,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 116064, strain column 10, stress column 11; maximum over 126540 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
8
- max_stress,exp_fc4984656462,197.459,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 64453, strain column 19, stress column 20; maximum over 64459 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
9
- max_stress,exp_3fed9f8f07c9,61.4139,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 5162, strain column 1, stress column 2; maximum over 5162 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
10
- max_stress,exp_aaa9e2c6f5b9,148.722,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 13882, strain column 4, stress column 5; maximum over 13881 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
11
- strain_at_max,exp_54b8fc8b3571,3.58553,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 70960, strain column 7, stress column 8; maximum over 81959 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
12
- strain_at_max,exp_3aa8fedf978b,6.20628,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 118267, strain column 25, stress column 26; maximum over 156878 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
13
- strain_at_max,exp_2d9d2ad06f6e,5.59208,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 112839, strain column 13, stress column 14; maximum over 186708 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
14
- strain_at_max,exp_8cfb0d14a259,6.15589,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 119513, strain column 16, stress column 17; maximum over 141872 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
15
- strain_at_max,exp_d53afffea8ad,0.67426,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 13882, strain column 4, stress column 5; maximum over 13881 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
16
- strain_at_max,exp_69a2a5342ec6,0.26561,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 5162, strain column 1, stress column 2; maximum over 5162 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
17
- strain_at_max,exp_eb05cc2a9dc8,1.57783,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 7608, strain column 22, stress column 23; maximum over 7607 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
18
- strain_at_max,exp_7b741589c10a,6.31575,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 116064, strain column 10, stress column 11; maximum over 126540 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
19
- strain_at_max,exp_4369fc61d66d,3.19442,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 64453, strain column 19, stress column 20; maximum over 64459 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
 
1
  group_id,item_id,mean,n,uncertainty_status,uncertainty_scale,uncertainty_basis,quality_status,warnings,source_location,aggregation
2
+ max_stress,exp_003fe25c3f91,203.947,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 112839, strain column 13, stress column 14; maximum over 186708 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
3
+ max_stress,exp_224a125b3d5a,197.459,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 64453, strain column 19, stress column 20; maximum over 64459 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
4
+ max_stress,exp_63b9d2553a78,184.302,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 116064, strain column 10, stress column 11; maximum over 126540 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
5
+ max_stress,exp_672208100095,208.152,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 119513, strain column 16, stress column 17; maximum over 141872 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
6
+ max_stress,exp_7c4f7caec6cf,148.722,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 13882, strain column 4, stress column 5; maximum over 13881 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
7
+ max_stress,exp_c339123d0b13,255.958,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 118267, strain column 25, stress column 26; maximum over 156878 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
8
+ max_stress,exp_cf10a9056f26,61.4139,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 5162, strain column 1, stress column 2; maximum over 5162 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
9
+ max_stress,exp_ea54530b4221,225.883,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 70960, strain column 7, stress column 8; maximum over 81959 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
10
+ max_stress,exp_ff5031900fc6,234.551,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 7608, strain column 22, stress column 23; maximum over 7607 paired curve points",Maximum of all finite paired stress samples in the original curve; no interpolation.
11
+ strain_at_max,exp_0a7a7c7f2b75,5.59208,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 112839, strain column 13, stress column 14; maximum over 186708 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
12
+ strain_at_max,exp_235fb0dc6cd6,3.19442,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 64453, strain column 19, stress column 20; maximum over 64459 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
13
+ strain_at_max,exp_351f1bf17bc1,3.58553,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 70960, strain column 7, stress column 8; maximum over 81959 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
14
+ strain_at_max,exp_3df219ed4754,0.26561,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 5162, strain column 1, stress column 2; maximum over 5162 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
15
+ strain_at_max,exp_6cc95c89ecb5,6.20628,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 118267, strain column 25, stress column 26; maximum over 156878 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
16
+ strain_at_max,exp_7ae22fdad3ab,6.15589,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 119513, strain column 16, stress column 17; maximum over 141872 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
17
+ strain_at_max,exp_84f36130f06f,0.67426,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 13882, strain column 4, stress column 5; maximum over 13881 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
18
+ strain_at_max,exp_d1a81b569dde,6.31575,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 116064, strain column 10, stress column 11; maximum over 126540 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
19
+ strain_at_max,exp_fb7f19041d5a,1.57783,,unknown,,No reliable independent difference uncertainty is supplied for this selected endpoint. Serial curve points are not independent sample replicates.,retained_with_flag,"[""Unknown independent uncertainty; use reported values without inventing zero uncertainty.""]","Original-SourceData.xlsx 'Figure 1b' row 7608, strain column 22, stress column 23; maximum over 7607 paired curve points",Engineering strain at the first occurrence of the original curve's maximum stress; no interpolation.
tasks/agcute-bending-endpoints-ranking/tests/original_grade.py DELETED
@@ -1,423 +0,0 @@
1
- """Pipeline v16: equal pairs within groups; one task-level random-95 threshold."""
2
-
3
- import argparse
4
- import errno
5
- import hashlib
6
- import itertools
7
- import json
8
- import math
9
- import os
10
- from pathlib import Path
11
- import stat
12
-
13
-
14
- class TaskError(ValueError):
15
- pass
16
-
17
-
18
- class SubmissionError(ValueError):
19
- pass
20
-
21
-
22
- def canonical_sha(value):
23
- return hashlib.sha256(json.dumps(value, sort_keys=True, separators=(",", ":"),
24
- ensure_ascii=False, allow_nan=False).encode()).hexdigest()
25
-
26
-
27
- def scoring_bindings(contract, answers):
28
- return {
29
- "contract_sha256": canonical_sha(contract),
30
- "answers_sha256": canonical_sha(answers),
31
- "relations_sha256": canonical_sha([{"group_id": g["group_id"], "pairs": g["pairs"]}
32
- for g in answers["groups"]]),
33
- "weights_sha256": canonical_sha([{k: g[k] for k in ("group_id", "group_weight")}
34
- for g in contract["groups"]]),
35
- "grader_sha256": hashlib.sha256(Path(__file__).read_bytes()).hexdigest(),
36
- }
37
-
38
-
39
- def validate_baseline(contract, answers, baseline):
40
- if not isinstance(baseline, dict):
41
- raise TaskError("random baseline must be an object")
42
- q = finite(baseline.get("q"), "random_q95", 0.5)
43
- if q >= 1:
44
- raise TaskError("random_q95 must be below 1")
45
- if contract.get("random_quantile") != 0.95:
46
- raise TaskError("invalid random quantile")
47
- if baseline.get("bindings") != scoring_bindings(contract, answers):
48
- raise TaskError("random baseline hash mismatch")
49
- return q
50
-
51
-
52
- def reward_from_accuracy(accuracy, q):
53
- accuracy = finite(accuracy, "accuracy", 0)
54
- q = finite(q, "random_q95", 0.5)
55
- if accuracy > 1 or q >= 1:
56
- raise TaskError("invalid accuracy or random threshold")
57
- if accuracy == 1:
58
- return 1.0
59
- if accuracy <= q + 1e-12:
60
- return 0.0
61
- return min(1.0, max(0.0, (accuracy - q) / (1 - q)))
62
-
63
-
64
- def aggregate_accuracy(details):
65
- # Preserve exact endpoints despite representational roundoff in weights.
66
- if all(g["group_pair_accuracy"] == 1 for g in details):
67
- return 1.0
68
- return math.fsum(g["group_pair_accuracy"] * g["group_weight"] for g in details)
69
-
70
-
71
- def unique_object(pairs):
72
- result = {}
73
- for key, value in pairs:
74
- if key in result:
75
- raise SubmissionError("duplicate_json_key")
76
- result[key] = value
77
- return result
78
-
79
-
80
- def load_json(raw):
81
- def reject_constant(value):
82
- raise SubmissionError("nonfinite_json_constant")
83
- return json.loads(raw.decode("utf-8"), object_pairs_hook=unique_object,
84
- parse_constant=reject_constant)
85
-
86
-
87
- def finite(value, name, minimum=None):
88
- if isinstance(value, bool) or not isinstance(value, (int, float)):
89
- raise TaskError(name + " is not numeric")
90
- if not math.isfinite(value) or (minimum is not None and value < minimum):
91
- raise TaskError(name + " is out of range")
92
- return float(value)
93
-
94
-
95
- def difference_se(sem_i, sem_j, method):
96
- if method == "sem_sum_upper_bound":
97
- return sem_i + sem_j
98
- if method == "independent_sem_quadrature":
99
- return math.hypot(sem_i, sem_j)
100
- raise TaskError("unsupported difference standard error method")
101
-
102
-
103
- def validate_task(contract, answers):
104
- if not isinstance(contract, dict) or not isinstance(answers, dict):
105
- raise TaskError("contract and answers must be objects")
106
- if contract.get("scoring_method") != "equal_pair_inversion_random95":
107
- raise TaskError("unsupported scoring method")
108
- if contract.get("unknown_uncertainty_policy") != "use_reported_value":
109
- raise TaskError("unsupported unknown-uncertainty policy")
110
- for key in ("task_id", "task_version"):
111
- if not isinstance(contract.get(key), str) or answers.get(key) != contract[key]:
112
- raise TaskError("invalid task identity")
113
- limit = contract.get("max_submission_bytes")
114
- if type(limit) is not int or not 0 < limit <= 1048576:
115
- raise TaskError("invalid submission size limit")
116
- minimum_conditions = contract.get("minimum_conditions_per_group", 2)
117
- if type(minimum_conditions) is not int or minimum_conditions < 2:
118
- raise TaskError("invalid minimum conditions")
119
- groups = contract.get("groups")
120
- hidden_groups = answers.get("groups")
121
- if not isinstance(groups, list) or not groups or not isinstance(hidden_groups, list):
122
- raise TaskError("missing groups")
123
- if len(groups) != len(hidden_groups):
124
- raise TaskError("hidden group count mismatch")
125
- hidden_map = {}
126
- for hidden in hidden_groups:
127
- if not isinstance(hidden, dict) or not isinstance(hidden.get("group_id"), str):
128
- raise TaskError("invalid hidden group")
129
- if hidden["group_id"] in hidden_map:
130
- raise TaskError("duplicate hidden group")
131
- hidden_map[hidden["group_id"]] = hidden
132
- validated, group_ids, all_ids = [], set(), set()
133
- for group in groups:
134
- if not isinstance(group, dict):
135
- raise TaskError("invalid contract group")
136
- gid = group.get("group_id")
137
- if not isinstance(gid, str) or not gid or gid in group_ids:
138
- raise TaskError("invalid group id")
139
- group_ids.add(gid)
140
- ids = group.get("item_ids")
141
- if (not isinstance(ids, list) or len(ids) < minimum_conditions
142
- or any(not isinstance(item, str) or not item for item in ids)
143
- or len(ids) != len(set(ids)) or all_ids.intersection(ids)):
144
- raise TaskError("invalid item ids")
145
- all_ids.update(ids)
146
- if group.get("direction") != "ascending":
147
- raise TaskError("only ascending order is supported")
148
- group_weight = finite(group.get("group_weight"), "group_weight", 0)
149
- if group_weight == 0:
150
- raise TaskError("zero group weight")
151
- k = finite(group.get("k"), "k", 0)
152
- epsilon = finite(group.get("epsilon"), "epsilon", 0)
153
- method = group.get("difference_standard_error")
154
- if method not in ("sem_sum_upper_bound", "independent_sem_quadrature", "unknown"):
155
- raise TaskError("unsupported difference standard error method")
156
- if group.get("uncertainty_model") not in ("optional_reported_uncertainty",
157
- "reported_sd_with_source_bounds", "independent_replicate_mean", "replicate_sd_upper_envelope", "unknown_replicate_hierarchy"):
158
- raise TaskError("unsupported marginal uncertainty model")
159
- hidden = hidden_map.get(gid)
160
- if hidden is None or hidden.get("quality_status") != "passed":
161
- raise TaskError("missing or invalid hidden group")
162
- items = hidden.get("items")
163
- if not isinstance(items, list) or len(items) != len(ids):
164
- raise TaskError("hidden item count mismatch")
165
- item_map = {}
166
- for item in items:
167
- if not isinstance(item, dict) or item.get("item_id") not in ids:
168
- raise TaskError("invalid hidden item")
169
- item_id = item["item_id"]
170
- if item_id in item_map or item.get("quality_status") not in ("retained", "retained_with_flag"):
171
- raise TaskError("duplicate or invalid hidden item")
172
- mean = finite(item.get("mean"), "mean")
173
- n = item.get("n")
174
- if n is not None and (type(n) is not int or n < 1):
175
- raise TaskError("invalid experimental replicate count")
176
- uncertainty_status = item.get("uncertainty_status")
177
- if uncertainty_status not in ("known", "unknown"):
178
- raise TaskError("invalid uncertainty status")
179
- uncertainty_model = group.get("uncertainty_model", "independent_replicate_mean")
180
- if uncertainty_model == "reported_sd_with_source_bounds":
181
- lo = finite(item.get("mean_lower"), "mean_lower")
182
- hi = finite(item.get("mean_upper"), "mean_upper")
183
- if lo > hi or not lo - 1e-12 <= mean <= hi + 1e-12:
184
- raise TaskError("inconsistent reported mean interval")
185
- if uncertainty_status == "unknown":
186
- if item.get("uncertainty_scale") is not None:
187
- raise TaskError("unknown uncertainty must be null")
188
- sem = None
189
- elif uncertainty_model == "optional_reported_uncertainty":
190
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
191
- if not item.get("uncertainty_basis"):
192
- raise TaskError("known uncertainty needs a stated source basis")
193
- elif uncertainty_model == "reported_sd_with_source_bounds":
194
- if n is None or n < 2:
195
- raise TaskError("reported SD needs repeat evidence")
196
- lower_sd = finite(item.get("sd_lower"), "sd_lower", 0)
197
- upper_sd = finite(item.get("sd_upper"), "sd_upper", 0)
198
- if lower_sd > upper_sd:
199
- raise TaskError("inconsistent reported SD interval")
200
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
201
- expected_scale = upper_sd + (hi - lo) / 2
202
- if not math.isclose(sem, expected_scale, rel_tol=1e-12, abs_tol=1e-12):
203
- raise TaskError("inconsistent source-bound uncertainty")
204
- elif uncertainty_model in ("independent_replicate_mean", "replicate_sd_upper_envelope"):
205
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
206
- sd = finite(item.get("sample_sd"), "sample_sd", 0)
207
- if n is None or n < 2:
208
- raise TaskError("known SD needs repeat evidence")
209
- expected_sem = sd / math.sqrt(n) if uncertainty_model == "independent_replicate_mean" else sd
210
- if not math.isclose(sem, expected_sem, rel_tol=1e-12, abs_tol=1e-12):
211
- raise TaskError("inconsistent hidden uncertainty")
212
- else:
213
- raise TaskError("unsupported marginal uncertainty model")
214
- item_map[item_id] = {"mean": mean, "sem": sem, "uncertainty_status": uncertainty_status}
215
- stored = hidden.get("pairs")
216
- if not isinstance(stored, list) or len(stored) != len(ids) * (len(ids) - 1) // 2:
217
- raise TaskError("invalid hidden pair count")
218
- stored_map = {}
219
- for pair in stored:
220
- if not isinstance(pair, dict):
221
- raise TaskError("invalid hidden pair")
222
- key = (pair.get("item_i"), pair.get("item_j"))
223
- if key in stored_map:
224
- raise TaskError("duplicate hidden pair")
225
- stored_map[key] = pair
226
- pairs = []
227
- for item_i, item_j in itertools.combinations(ids, 2):
228
- left, right = item_map[item_i], item_map[item_j]
229
- difference = left["mean"] - right["mean"]
230
- known = left["uncertainty_status"] == right["uncertainty_status"] == "known"
231
- se = difference_se(left["sem"], right["sem"], method) if known else None
232
- tolerance = k * se + epsilon if known else epsilon
233
- comparable = int(abs(difference) > tolerance)
234
- basis = "uncertainty_adjusted" if known else "reported_value"
235
- if not math.isfinite(difference) or not math.isfinite(tolerance) or (known and not math.isfinite(se)):
236
- raise TaskError("nonfinite pair arithmetic")
237
- pair = stored_map.get((item_i, item_j))
238
- if pair is None:
239
- raise TaskError("missing hidden pair")
240
- for name, expected, minimum in (("mean_difference", difference, None),):
241
- actual = finite(pair.get(name), name, minimum)
242
- if not math.isclose(actual, expected, rel_tol=1e-12, abs_tol=1e-10):
243
- raise TaskError("inconsistent hidden " + name)
244
- if known:
245
- if not math.isclose(finite(pair.get("difference_se"), "difference_se", 0), se,
246
- rel_tol=1e-12, abs_tol=1e-10):
247
- raise TaskError("inconsistent pair uncertainty")
248
- elif pair.get("difference_se") is not None:
249
- raise TaskError("unknown pair uncertainty must be null")
250
- if type(pair.get("comparable")) is not int or pair["comparable"] != comparable:
251
- raise TaskError("inconsistent comparable flag")
252
- if pair.get("uncertainty_status") != ("known" if known else "unknown"):
253
- raise TaskError("inconsistent pair uncertainty status")
254
- if pair.get("comparison_basis") != basis:
255
- raise TaskError("inconsistent comparison basis")
256
- pairs.append({"item_i": item_i, "item_j": item_j,
257
- "mean_difference": difference, "difference_se": se,
258
- "uncertainty_status": "known" if known else "unknown",
259
- "comparison_basis": basis, "comparable": comparable})
260
- total = sum(pair["comparable"] for pair in pairs)
261
- if total <= 0:
262
- raise TaskError("group has no scoreable pair under the declared rule")
263
- if type(hidden.get("comparable_pairs")) is not int or hidden["comparable_pairs"] != total:
264
- raise TaskError("inconsistent comparable pair count")
265
- validated.append({"group_id": gid, "item_ids": ids, "group_weight": group_weight,
266
- "pairs": pairs, "comparable_pairs": total})
267
- if set(hidden_map) != group_ids:
268
- raise TaskError("unknown hidden group")
269
- if not math.isclose(math.fsum(g["group_weight"] for g in validated), 1.0,
270
- rel_tol=0, abs_tol=1e-12):
271
- raise TaskError("group weights must sum to one")
272
- return validated
273
-
274
-
275
- def zero_report(contract, reason):
276
- return {"task_id": contract["task_id"], "task_version": contract["task_version"],
277
- "ranking_reward": 0.0, "submission_status": reason, "groups": [],
278
- "task_pair_accuracy": 0.0, "aggregation_zero_due_to_format": True,
279
- "official_reward": None, "review_status": "not_reviewed", "evaluation_status": "completed"}
280
-
281
-
282
- def grade_submission(contract, answers, submission, baseline):
283
- groups = validate_task(contract, answers)
284
- q = validate_baseline(contract, answers, baseline)
285
- if not isinstance(submission, dict):
286
- return zero_report(contract, "invalid_top_level_structure")
287
- if any(submission.get(key) != contract[key] for key in ("task_id", "task_version")):
288
- return zero_report(contract, "task_identity_mismatch")
289
- entries = submission.get("rankings")
290
- if not isinstance(entries, list):
291
- return zero_report(contract, "rankings_not_array")
292
- known = {g["group_id"] for g in groups}
293
- by_group = {}
294
- for entry in entries:
295
- if not isinstance(entry, dict) or not isinstance(entry.get("group_id"), str):
296
- return zero_report(contract, "unassignable_group_entry")
297
- if entry["group_id"] not in known:
298
- return zero_report(contract, "unknown_group_id")
299
- by_group.setdefault(entry["group_id"], []).append(entry)
300
- details = []
301
- for group in groups:
302
- gid = group["group_id"]
303
- matches = by_group.get(gid, [])
304
- item = {"group_id": gid, "group_weight": group["group_weight"],
305
- "group_pair_accuracy": 0.0, "status": "missing_group", "comparable_pairs": group["comparable_pairs"],
306
- "total_pairs": len(group["pairs"]), "item_count": len(group["item_ids"]),
307
- "coverage": group["comparable_pairs"] / len(group["pairs"])}
308
- if len(matches) > 1:
309
- item["status"] = "duplicate_group"
310
- elif len(matches) == 1:
311
- entry = matches[0]
312
- order = entry.get("order")
313
- if (not isinstance(order, list) or any(not isinstance(x, str) for x in order)
314
- or len(order) != len(group["item_ids"])
315
- or len(order) != len(set(order)) or set(order) != set(group["item_ids"])):
316
- item["status"] = "invalid_permutation"
317
- else:
318
- positions = {item_id: i for i, item_id in enumerate(order)}
319
- pair_details = []
320
- for pair in group["pairs"]:
321
- wrong = ((positions[pair["item_i"]] - positions[pair["item_j"]])
322
- * pair["mean_difference"]) < 0
323
- pair_details.append({**pair, "reversed": wrong,
324
- "counted_inversion": int(wrong and pair["comparable"] == 1)})
325
- lost = sum(pair["counted_inversion"] for pair in pair_details)
326
- total = group["comparable_pairs"]
327
- reward = 1.0 - lost / total
328
- item.update(status="valid", group_pair_accuracy=max(0.0, min(1.0, reward)),
329
- order=order, inversion_pairs=lost, correct_pairs=total-lost,
330
- comparable_pairs=total, total_pairs=len(pair_details),
331
- coverage=total/len(pair_details),
332
- reported_value_counted_pairs=sum(p["comparable"] for p in pair_details
333
- if p["comparison_basis"] == "reported_value"),
334
- uncertainty_adjusted_counted_pairs=sum(p["comparable"] for p in pair_details
335
- if p["comparison_basis"] == "uncertainty_adjusted"),
336
- all_reported_value_unequal_pairs=sum(p["mean_difference"] != 0 for p in pair_details),
337
- all_reported_value_correct_pairs=sum(p["mean_difference"] != 0 and not p["reversed"]
338
- for p in pair_details),
339
- pairs=pair_details)
340
- details.append(item)
341
- return {"task_id": contract["task_id"], "task_version": contract["task_version"],
342
- "task_pair_accuracy": aggregate_accuracy(details),
343
- "random_q95": q,
344
- "ranking_reward": reward_from_accuracy(aggregate_accuracy(details), q),
345
- "submission_status": "valid" if all(g["status"] == "valid" for g in details) else "group_errors",
346
- "groups": details, "official_reward": None,
347
- "review_status": "not_reviewed", "evaluation_status": "completed"}
348
-
349
-
350
- def read_regular(path, limit):
351
- try:
352
- fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
353
- try:
354
- info = os.fstat(fd)
355
- if not stat.S_ISREG(info.st_mode):
356
- raise SubmissionError("not_regular_file")
357
- if info.st_size > limit:
358
- raise SubmissionError("submission_too_large")
359
- with os.fdopen(fd, "rb", closefd=False) as stream:
360
- raw = stream.read(limit + 1)
361
- if len(raw) > limit:
362
- raise SubmissionError("submission_too_large")
363
- return raw
364
- finally:
365
- os.close(fd)
366
- except OSError as error:
367
- if error.errno in (errno.ENOENT, errno.ELOOP, errno.EISDIR, errno.ENOTDIR,
368
- errno.ENXIO, errno.EACCES, errno.EPERM):
369
- raise SubmissionError("missing_or_unreadable_submission") from error
370
- raise
371
-
372
-
373
- def grade_file(contract, answers, path, baseline):
374
- validate_task(contract, answers)
375
- q = validate_baseline(contract, answers, baseline)
376
- try:
377
- raw = read_regular(path, contract["max_submission_bytes"])
378
- submission = load_json(raw)
379
- except (SubmissionError, UnicodeError, json.JSONDecodeError, RecursionError) as error:
380
- report = zero_report(contract, str(error) or type(error).__name__)
381
- report["random_q95"] = q
382
- return report
383
- report = grade_submission(contract, answers, submission, baseline)
384
- report["random_q95"] = q
385
- report["submission_sha256"] = hashlib.sha256(raw).hexdigest()
386
- return report
387
-
388
-
389
- def dump(path, value):
390
- path.write_text(json.dumps(value, indent=2, allow_nan=False) + "\n", encoding="utf-8")
391
-
392
-
393
- def main():
394
- parser = argparse.ArgumentParser()
395
- parser.add_argument("--contract", type=Path, default=Path("/tests/contract.json"))
396
- parser.add_argument("--answers", type=Path, default=Path("/tests/answers.json"))
397
- parser.add_argument("--baseline", type=Path, default=Path("/tests/random_baseline.json"))
398
- parser.add_argument("--submission", type=Path, default=Path("/logs/artifacts/rankings.json"))
399
- parser.add_argument("--output-dir", type=Path, default=Path("/logs/verifier"))
400
- args = parser.parse_args()
401
- args.output_dir.mkdir(parents=True, exist_ok=True)
402
- for name in ("reward.txt", "reward.json", "report.json", "evaluator_error.json"):
403
- (args.output_dir / name).unlink(missing_ok=True)
404
- try:
405
- contract = load_json(args.contract.read_bytes())
406
- answers = load_json(args.answers.read_bytes())
407
- baseline = load_json(args.baseline.read_bytes())
408
- report = grade_file(contract, answers, args.submission, baseline)
409
- except (TaskError, SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
410
- dump(args.output_dir / "evaluator_error.json", {"status": "task_error", "error": str(error)})
411
- raise SystemExit(2) from error
412
- dump(args.output_dir / "report.json", report)
413
- (args.output_dir / "reward.txt").write_text(str(report["ranking_reward"]) + "\n", encoding="utf-8")
414
- print(json.dumps({"ranking_reward": report["ranking_reward"],
415
- "submission_status": report["submission_status"],
416
- "task_pair_accuracy": report["task_pair_accuracy"],
417
- "random_q95": report["random_q95"],
418
- "aggregation_zero_due_to_format": report.get("aggregation_zero_due_to_format", False),
419
- "official_reward": None, "review_status": "not_reviewed", "evaluation_status": "completed"}))
420
-
421
-
422
- if __name__ == "__main__":
423
- main()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tasks/agcute-bending-endpoints-ranking/tests/random_baseline.json DELETED
@@ -1,66 +0,0 @@
1
- {
2
- "method": "uniform_complete_permutations",
3
- "generator": "numpy.random.PCG64",
4
- "numpy_version": "2.4.4",
5
- "quantile": 0.95,
6
- "quantile_method": "higher",
7
- "calibration_count": 200000,
8
- "validation_count": 200000,
9
- "calibration_seed": 17436290277762341355,
10
- "validation_seed": 7360250039651054117,
11
- "q": 0.6527777777777778,
12
- "task_id": "agcute-bending-endpoints-ranking",
13
- "task_version": "1.0.0",
14
- "group_order": [
15
- "max_stress",
16
- "strain_at_max"
17
- ],
18
- "item_order": {
19
- "max_stress": [
20
- "exp_65960918c5e1",
21
- "exp_0b361eedc092",
22
- "exp_0a801637748b",
23
- "exp_39fd1eed6b1c",
24
- "exp_365507fb236c",
25
- "exp_3c3c8522da9d",
26
- "exp_fc4984656462",
27
- "exp_3fed9f8f07c9",
28
- "exp_aaa9e2c6f5b9"
29
- ],
30
- "strain_at_max": [
31
- "exp_54b8fc8b3571",
32
- "exp_3aa8fedf978b",
33
- "exp_2d9d2ad06f6e",
34
- "exp_8cfb0d14a259",
35
- "exp_d53afffea8ad",
36
- "exp_69a2a5342ec6",
37
- "exp_eb05cc2a9dc8",
38
- "exp_7b741589c10a",
39
- "exp_4369fc61d66d"
40
- ]
41
- },
42
- "calibration_code_sha256": "cbb437c19ae80d85f28374fe62c71e8052975ddc48c447fe89b9fb9913e8c6b1",
43
- "relations_sha256": "c6be6dea1116a1cb8b9f113166040f1e0f907cb91946375bfcbe2578804943bc",
44
- "weights_sha256": "e2f8f70a28b53aadf0d15ba87db844c29814b1f2511f1fbedb4f15da3007c20e",
45
- "validation": {
46
- "task_pair_accuracy": {
47
- "mean": 0.4998954166666667,
48
- "mean_standard_error": 0.00021030780229321986
49
- },
50
- "task_spearman": {
51
- "mean": -1.5541666666666494e-05,
52
- "mean_standard_error": 0.0005583706707422961
53
- },
54
- "ranking_reward": {
55
- "mean": 0.005649400000000001,
56
- "mean_standard_error": 7.235473429042408e-05
57
- },
58
- "zero_reward_fraction": 0.95489
59
- },
60
- "bindings": {
61
- "contract_sha256": "f584d2b0ec67e9f047bc763fce2e0afc42a9fe4a2676db8c826dfec8dabc4ae2",
62
- "answers_sha256": "beb49bf263eccb3f24b4bf2f0eddbc0d255a358b34418a20e858708ccf304e19",
63
- "grader_sha256": "82c9d3b9c1fa5c2e2a39852e315df9e04714dee2ffa9eb64373014d8b4aeb381",
64
- "auxiliary_grader_sha256": "b91d5af76c52dca0cb7195777a0cbea690a91af15f6bdbb0b13714e03ab45492"
65
- }
66
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tasks/agcute-bending-endpoints-ranking/tests/scoring_hashes.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "answers.json": "ff2ac247617efc1e5aa4f7e4fc66cb30151772282aae845a283355a09f6af78e",
3
+ "contract.json": "87fde63e7da41c31c00741b678b831d8e3614ca0a6e065ebc5cf69ac4bcbe816",
4
+ "grade.py": "0c25490ea9639fd4e73fde4741f4b4e02bdbd42962779348694e50e33a121550"
5
+ }
tasks/alkene-transfer-conditions-ranking/environment/public/contract.json CHANGED
@@ -1,72 +1,59 @@
1
  {
2
  "task_id": "alkene-transfer-conditions-ranking",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "g1",
7
  "label": "4-Phenyl-1-butene cross-metathesis adduct yield",
8
  "item_ids": [
9
- "exp_41bd98dba742",
10
- "exp_484d7efaebd5",
11
- "exp_389c830f7e89",
12
- "exp_dd127eb2b1f5",
13
- "exp_8545262bd121",
14
- "exp_369ffbca2e28",
15
- "exp_cac98fcd26c1"
16
  ],
17
  "direction": "ascending",
18
  "group_weight": 0.3333333333333333,
19
- "k": 2,
20
- "epsilon": 0,
21
- "uncertainty_model": "optional_reported_uncertainty",
22
- "difference_standard_error": "sem_sum_upper_bound",
23
  "endpoint": {
24
  "name": "product 1H NMR yield",
25
  "unit": "%",
26
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
27
  }
28
  },
29
  {
30
  "group_id": "g2",
31
  "label": "Rotenone cross-metathesis adduct yield",
32
  "item_ids": [
33
- "exp_066e33c896b9",
34
- "exp_fba92a14e6bb",
35
- "exp_9279aa7cd389",
36
- "exp_8e5983854cd1",
37
- "exp_343b2ac1c9a7",
38
- "exp_e01ea673fa5d",
39
- "exp_ea300526f709"
40
  ],
41
  "direction": "ascending",
42
  "group_weight": 0.3333333333333333,
43
- "k": 2,
44
- "epsilon": 0,
45
- "uncertainty_model": "optional_reported_uncertainty",
46
- "difference_standard_error": "sem_sum_upper_bound",
47
  "endpoint": {
48
  "name": "product 1H NMR yield",
49
  "unit": "%",
50
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
51
  }
52
  },
53
  {
54
  "group_id": "g3",
55
  "label": "Combined alkene yield from E/Z latent precursors",
56
  "item_ids": [
57
- "exp_529d2754aa6a",
58
- "exp_94c775c85236"
59
  ],
60
  "direction": "ascending",
61
  "group_weight": 0.3333333333333333,
62
- "k": 2,
63
- "epsilon": 0,
64
- "uncertainty_model": "optional_reported_uncertainty",
65
- "difference_standard_error": "sem_sum_upper_bound",
66
  "endpoint": {
67
  "name": "product 1H NMR yield",
68
  "unit": "%",
69
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
70
  }
71
  }
72
  ],
@@ -79,18 +66,9 @@
79
  "spearman_tie_method": "average",
80
  "spearman_undefined_policy": "task_error",
81
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
82
- "secondary_scoring_method": "equal_pair_inversion_random95",
83
- "ranking_reward_range": [
84
- 0,
85
- 1
86
- ],
87
- "random_quantile": 0.95,
88
- "unknown_uncertainty_policy": "use_reported_value",
89
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
90
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
91
  "format_errors": {
92
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
93
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
94
  },
95
  "max_submission_bytes": 1048576,
96
  "minimum_conditions_per_group": 2,
@@ -107,5 +85,14 @@
107
  "report_path": "/logs/artifacts/report.md",
108
  "sources_path": "/logs/artifacts/sources.json",
109
  "analysis_path": "/logs/artifacts/analysis/",
110
- "forbidden_result_scope": "Measured outcomes, yields, rankings, equivalent endpoints and result-fitted predictors for alkene transfer-reagent and catalyst conditions in the stated material and reaction system are prohibited."
 
 
 
 
 
 
 
 
 
111
  }
 
1
  {
2
  "task_id": "alkene-transfer-conditions-ranking",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "g1",
7
  "label": "4-Phenyl-1-butene cross-metathesis adduct yield",
8
  "item_ids": [
9
+ "exp_6c7c207a3e35",
10
+ "exp_777d484cc14b",
11
+ "exp_7ee0f9400be1",
12
+ "exp_b25afc3d08f5",
13
+ "exp_ecccb58cf20b",
14
+ "exp_f6a85f2013bc"
 
15
  ],
16
  "direction": "ascending",
17
  "group_weight": 0.3333333333333333,
 
 
 
 
18
  "endpoint": {
19
  "name": "product 1H NMR yield",
20
  "unit": "%",
21
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
22
  }
23
  },
24
  {
25
  "group_id": "g2",
26
  "label": "Rotenone cross-metathesis adduct yield",
27
  "item_ids": [
28
+ "exp_039643486694",
29
+ "exp_064708d49167",
30
+ "exp_121a21fd986b",
31
+ "exp_3c5d2113087d",
32
+ "exp_5fe231ef098e",
33
+ "exp_a3d319499eac",
34
+ "exp_dddc25cdd5a3"
35
  ],
36
  "direction": "ascending",
37
  "group_weight": 0.3333333333333333,
 
 
 
 
38
  "endpoint": {
39
  "name": "product 1H NMR yield",
40
  "unit": "%",
41
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
42
  }
43
  },
44
  {
45
  "group_id": "g3",
46
  "label": "Combined alkene yield from E/Z latent precursors",
47
  "item_ids": [
48
+ "exp_b167406216a1",
49
+ "exp_b81cc665557a"
50
  ],
51
  "direction": "ascending",
52
  "group_weight": 0.3333333333333333,
 
 
 
 
53
  "endpoint": {
54
  "name": "product 1H NMR yield",
55
  "unit": "%",
56
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
57
  }
58
  }
59
  ],
 
66
  "spearman_tie_method": "average",
67
  "spearman_undefined_policy": "task_error",
68
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
69
  "format_errors": {
70
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
71
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
72
  },
73
  "max_submission_bytes": 1048576,
74
  "minimum_conditions_per_group": 2,
 
85
  "report_path": "/logs/artifacts/report.md",
86
  "sources_path": "/logs/artifacts/sources.json",
87
  "analysis_path": "/logs/artifacts/analysis/",
88
+ "forbidden_result_scope": "Measured outcomes, yields, rankings, equivalent endpoints and result-fitted predictors for alkene transfer-reagent and catalyst conditions in the stated material and reaction system are prohibited.",
89
+ "supporting_artifact_review": {
90
+ "policy_stop_marker": "SPU_POLICY_STOP",
91
+ "required_files": [
92
+ "report.md",
93
+ "sources.json"
94
+ ],
95
+ "required_directory": "analysis",
96
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
97
+ }
98
  }
tasks/alkene-transfer-conditions-ranking/environment/public/experiment.md CHANGED
@@ -6,21 +6,21 @@ Compare the listed experimental conditions for alkene transfer-reagent and catal
6
 
7
  | Field | Setting |
8
  | --- | --- |
9
- | material_handling | `Oven-dried glassware underN2 usingSchlenk techniques.All specified solvents are anhydrous.Commercial reagents used as supplied;one-carbon-transfer reagents stored in freezer underN2,numeric storage temperature unspecified.` |
10
- | reagents | `Transfer reagent1a:((allylsulfonyl)methoxy)(tert-butyl)dimethylsilane,SMILES C=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. Transfer reagent1b:tert-butyldimethyl(((3-methylbut-2-en-1-yl)sulfonyl)methoxy)silane,SMILES CC(C)=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. TBS means tert-butyldimethylsilyl.` |
11
- | endpoint_convention | `Rank measured product1H NMR yield within each comparison group. Cross-metathesis groups concern total latent adduct; the E/Z-reactant group concerns the combined alkene isomers after fragmentation. Do not combine these endpoints across groups.` |
12
 
13
  ## Procedure and measurement
14
 
15
  Combine each listed condition with the common procedure. Rank the specified measured endpoint separately within each comparison group.
16
 
17
- - `g1` — product 1H NMR yield (%): Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard.
18
- - `g2` — product 1H NMR yield (%): Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard.
19
- - `g3` — product 1H NMR yield (%): Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard.
20
 
21
  ## Item settings
22
 
23
- Expand each item in targets.csv or protocol.json together with the common conditions. Every selected item is listed using the same fields.
24
 
25
  ## Source limitations relevant to the definition
26
 
@@ -35,7 +35,7 @@ HG-II,G-II andG-I catalyst molecular structures.
35
 
36
  ### structures/rotenone_reaction.png
37
 
38
- Rotenone stereostructure,transfer reagents1a/1b,andcross-metathesis product.
39
 
40
  ![Experimental structure](structures/rotenone_reaction.png)
41
 
 
6
 
7
  | Field | Setting |
8
  | --- | --- |
9
+ | material_handling | `Oven-dried glassware under N2 using Schlenk techniques. All specified solvents are anhydrous. Commercial reagents used as supplied; one-carbon-transfer reagents stored in freezer under N2,numeric storage temperature unspecified.` |
10
+ | reagents | `Transfer reagent1a:((allylsulfonyl)methoxy)(tert-butyl)dimethylsilane, SMILES C=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. Transfer reagent1b:tert-butyldimethyl(((3-methylbut-2-en-1-yl)sulfonyl)methoxy)silane, SMILES CC(C)=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. TBS means tert-butyldimethylsilyl.` |
11
+ | endpoint_convention | `Rank measured product 1H NMR yield within each comparison group. Cross-metathesis groups concern total latent adduct; the E/Z-reactant group concerns the combined alkene isomers after fragmentation. Do not combine these endpoints across groups.` |
12
 
13
  ## Procedure and measurement
14
 
15
  Combine each listed condition with the common procedure. Rank the specified measured endpoint separately within each comparison group.
16
 
17
+ - `g1` — product 1H NMR yield (%): Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard.
18
+ - `g2` — product 1H NMR yield (%): Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard.
19
+ - `g3` — product 1H NMR yield (%): Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard.
20
 
21
  ## Item settings
22
 
23
+ Expand each item in targets.csv or protocol.json together with the common conditions.
24
 
25
  ## Source limitations relevant to the definition
26
 
 
35
 
36
  ### structures/rotenone_reaction.png
37
 
38
+ Rotenone stereostructure, transfer reagents1a/1b,and cross-metathesis product.
39
 
40
  ![Experimental structure](structures/rotenone_reaction.png)
41
 
tasks/alkene-transfer-conditions-ranking/environment/public/information_policy.md CHANGED
@@ -10,3 +10,5 @@ Forbidden examples:
10
  - Measured or inferred rankings calibrated on the same-type experimental outcomes.
11
 
12
  Use audited tools for all searches, downloads and code-initiated network requests. Check both requests and returned content before use. Do not circumvent access controls or ask another model/service for forbidden results. If a permitted query unexpectedly exposes prohibited external results, immediately stop prediction and further search; record only the triggering step, source and stopping reason in report.md without restating results. Do not retrieve or reconstruct specific experimental results from your own model memory; briefly disclose a noticed recollection in your report and continue using permitted inputs and reasoning.
 
 
 
10
  - Measured or inferred rankings calibrated on the same-type experimental outcomes.
11
 
12
  Use audited tools for all searches, downloads and code-initiated network requests. Check both requests and returned content before use. Do not circumvent access controls or ask another model/service for forbidden results. If a permitted query unexpectedly exposes prohibited external results, immediately stop prediction and further search; record only the triggering step, source and stopping reason in report.md without restating results. Do not retrieve or reconstruct specific experimental results from your own model memory; briefly disclose a noticed recollection in your report and continue using permitted inputs and reasoning.
13
+
14
+ For a stop under the information policy, put `SPU_POLICY_STOP` on the first line of `report.md` and record the triggering step, source and reason below it. Withdraw any provisional ranking.
tasks/alkene-transfer-conditions-ranking/environment/public/protocol.json CHANGED
@@ -1,50 +1,50 @@
1
  {
2
  "common_conditions": {
3
- "material_handling": "Oven-dried glassware underN2 usingSchlenk techniques.All specified solvents are anhydrous.Commercial reagents used as supplied;one-carbon-transfer reagents stored in freezer underN2,numeric storage temperature unspecified.",
4
- "reagents": "Transfer reagent1a:((allylsulfonyl)methoxy)(tert-butyl)dimethylsilane,SMILES C=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. Transfer reagent1b:tert-butyldimethyl(((3-methylbut-2-en-1-yl)sulfonyl)methoxy)silane,SMILES CC(C)=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. TBS means tert-butyldimethylsilyl.",
5
- "endpoint_convention": "Rank measured product1H NMR yield within each comparison group. Cross-metathesis groups concern total latent adduct; the E/Z-reactant group concerns the combined alkene isomers after fragmentation. Do not combine these endpoints across groups."
6
  },
7
  "measurement": {
8
  "name": "product 1H NMR yield",
9
  "unit": "%",
10
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
11
  },
12
  "measurements_by_group": {
13
  "g1": {
14
  "name": "product 1H NMR yield",
15
  "unit": "%",
16
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
17
  },
18
  "g2": {
19
  "name": "product 1H NMR yield",
20
  "unit": "%",
21
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
22
  },
23
  "g3": {
24
  "name": "product 1H NMR yield",
25
  "unit": "%",
26
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
27
  }
28
  },
29
  "measurement_rule": "Use measurements_by_group for each group. The measurement field is a default summary and does not override a group-specific endpoint.",
30
  "inheritance_rule": "common_conditions apply unless the same field is explicitly overridden by the item; item settings are literal experimental facts, not predicted outcomes.",
31
  "items": {
32
- "exp_41bd98dba742": {
33
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
34
  "transfer_reagent": "1a",
35
- "transfer_reagent_equiv": 4,
36
- "catalyst": "G-II as structures/metathesis_catalysts.png",
37
  "catalyst_mol_percent": 5,
38
  "solvent": "dichloromethane",
39
  "solvent_volume_mL": 1,
40
  "alkene_concentration_M": 0.1,
41
  "temperature_C": 40,
42
  "time_h": 8,
43
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
44
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
45
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
46
  },
47
- "exp_484d7efaebd5": {
48
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
49
  "transfer_reagent": "1a",
50
  "transfer_reagent_equiv": 2,
@@ -55,11 +55,11 @@
55
  "alkene_concentration_M": 0.1,
56
  "temperature_C": 40,
57
  "time_h": 8,
58
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
59
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
60
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
61
  },
62
- "exp_389c830f7e89": {
63
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
64
  "transfer_reagent": "1a",
65
  "transfer_reagent_equiv": 1,
@@ -70,11 +70,11 @@
70
  "alkene_concentration_M": 0.1,
71
  "temperature_C": 40,
72
  "time_h": 8,
73
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
74
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
75
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
76
  },
77
- "exp_dd127eb2b1f5": {
78
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
79
  "transfer_reagent": "1a",
80
  "transfer_reagent_equiv": 4,
@@ -85,89 +85,74 @@
85
  "alkene_concentration_M": 0.1,
86
  "temperature_C": 40,
87
  "time_h": 8,
88
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
89
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
90
- "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
91
- },
92
- "exp_8545262bd121": {
93
- "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
94
- "transfer_reagent": "1b",
95
- "transfer_reagent_equiv": 3,
96
- "catalyst": "HG-II as structures/metathesis_catalysts.png",
97
- "catalyst_mol_percent": 5,
98
- "solvent": "toluene",
99
- "solvent_volume_mL": 1,
100
- "alkene_concentration_M": 0.1,
101
- "temperature_C": 100,
102
- "time_h": 8,
103
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
104
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
105
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
106
  },
107
- "exp_369ffbca2e28": {
108
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
109
- "transfer_reagent": "1b",
110
- "transfer_reagent_equiv": 3,
111
- "catalyst": "HG-II as structures/metathesis_catalysts.png",
112
  "catalyst_mol_percent": 5,
113
  "solvent": "dichloromethane",
114
  "solvent_volume_mL": 1,
115
  "alkene_concentration_M": 0.1,
116
  "temperature_C": 40,
117
  "time_h": 8,
118
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
119
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
120
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
121
  },
122
- "exp_cac98fcd26c1": {
123
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
124
- "transfer_reagent": "1a",
125
  "transfer_reagent_equiv": 3,
126
  "catalyst": "HG-II as structures/metathesis_catalysts.png",
127
  "catalyst_mol_percent": 5,
128
- "solvent": "dichloromethane",
129
  "solvent_volume_mL": 1,
130
  "alkene_concentration_M": 0.1,
131
- "temperature_C": 40,
132
  "time_h": 8,
133
- "procedure": "5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
134
- "analysis": "Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
135
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
136
  },
137
- "exp_066e33c896b9": {
138
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
139
- "transfer_reagent": "1b",
140
- "transfer_reagent_equiv": 3,
141
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
142
  "catalyst_mol_percent": 20,
143
  "solvent": "1,2-dichloroethane",
144
  "solvent_volume_mL": 0.5,
145
  "alkene_concentration_M": 0.1,
146
  "temperature_C": 100,
147
- "time_h": 16,
148
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
149
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
150
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
151
  },
152
- "exp_fba92a14e6bb": {
153
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
154
- "transfer_reagent": "1a",
155
- "transfer_reagent_equiv": 10,
156
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
157
  "catalyst_mol_percent": 20,
158
  "solvent": "1,2-dichloroethane",
159
  "solvent_volume_mL": 0.5,
160
  "alkene_concentration_M": 0.1,
161
  "temperature_C": 100,
162
- "time_h": 5,
163
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
164
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
165
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
166
  },
167
- "exp_9279aa7cd389": {
168
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
169
  "transfer_reagent": "1b",
170
- "transfer_reagent_equiv": 2,
171
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
172
  "catalyst_mol_percent": 20,
173
  "solvent": "1,2-dichloroethane",
@@ -175,11 +160,11 @@
175
  "alkene_concentration_M": 0.1,
176
  "temperature_C": 100,
177
  "time_h": 16,
178
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
179
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
180
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
181
  },
182
- "exp_8e5983854cd1": {
183
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
184
  "transfer_reagent": "1a",
185
  "transfer_reagent_equiv": 10,
@@ -190,11 +175,11 @@
190
  "alkene_concentration_M": 0.1,
191
  "temperature_C": 100,
192
  "time_h": 5,
193
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
194
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
195
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
196
  },
197
- "exp_343b2ac1c9a7": {
198
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
199
  "transfer_reagent": "1a",
200
  "transfer_reagent_equiv": 10,
@@ -205,14 +190,14 @@
205
  "alkene_concentration_M": 0.1,
206
  "temperature_C": 100,
207
  "time_h": 16,
208
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
209
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
210
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
211
  },
212
- "exp_e01ea673fa5d": {
213
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
214
  "transfer_reagent": "1b",
215
- "transfer_reagent_equiv": 5,
216
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
217
  "catalyst_mol_percent": 20,
218
  "solvent": "1,2-dichloroethane",
@@ -220,11 +205,11 @@
220
  "alkene_concentration_M": 0.1,
221
  "temperature_C": 100,
222
  "time_h": 16,
223
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
224
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
225
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
226
  },
227
- "exp_ea300526f709": {
228
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
229
  "transfer_reagent": "1a",
230
  "transfer_reagent_equiv": 10,
@@ -235,32 +220,32 @@
235
  "alkene_concentration_M": 0.1,
236
  "temperature_C": 40,
237
  "time_h": 16,
238
- "procedure": "10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.",
239
- "analysis": "Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
240
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
241
  },
242
- "exp_529d2754aa6a": {
243
- "latent_precursor": "E-17,fullmolecularstructure in structures/latent_E.png,36.8mg0.100mmol",
244
  "citric_acid": "57.6mg0.300mmol3equiv",
245
  "fluoride": "Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv",
246
  "solvent": "0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",
247
  "substrate_concentration_M": 0.1,
248
  "temperature_C": 40,
249
  "time_h": 16,
250
- "procedure": "10mLmicrowavevial withmagneticbar.Evacuate/refillN2threecycles.Adddichloromethane,heatin oilbath.Thenaddcitricacid andTBAFtrihydrate dissolved inmethanol throughseptum;heatandstir16h.",
251
- "analysis": "Addmethylbenzoate12.6microlitre0.100mmol.Takecrudealiquot for1H NMR combinedproductyield;GC-MS separatelydeterminesisomerratio,butratiois notthe requestedendpoint.",
252
  "product": "CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation."
253
  },
254
- "exp_94c775c85236": {
255
- "latent_precursor": "Z-17,fullmolecularstructure in structures/latent_Z.png,36.8mg0.100mmol",
256
  "citric_acid": "57.6mg0.300mmol3equiv",
257
  "fluoride": "Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv",
258
  "solvent": "0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",
259
  "substrate_concentration_M": 0.1,
260
  "temperature_C": 40,
261
  "time_h": 16,
262
- "procedure": "10mLmicrowavevial withmagneticbar.Evacuate/refillN2threecycles.Adddichloromethane,heatin oilbath.Thenaddcitricacid andTBAFtrihydrate dissolved inmethanol throughseptum;heatandstir16h.",
263
- "analysis": "Addmethylbenzoate12.6microlitre0.100mmol.Takecrudealiquot for1H NMR combinedproductyield;GC-MS separatelydeterminesisomerratio,butratiois notthe requestedendpoint.",
264
  "product": "CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation."
265
  }
266
  }
 
1
  {
2
  "common_conditions": {
3
+ "material_handling": "Oven-dried glassware under N2 using Schlenk techniques. All specified solvents are anhydrous. Commercial reagents used as supplied; one-carbon-transfer reagents stored in freezer under N2,numeric storage temperature unspecified.",
4
+ "reagents": "Transfer reagent1a:((allylsulfonyl)methoxy)(tert-butyl)dimethylsilane, SMILES C=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. Transfer reagent1b:tert-butyldimethyl(((3-methylbut-2-en-1-yl)sulfonyl)methoxy)silane, SMILES CC(C)=CCS(=O)(=O)CO[Si](C)(C)C(C)(C)C. TBS means tert-butyldimethylsilyl.",
5
+ "endpoint_convention": "Rank measured product 1H NMR yield within each comparison group. Cross-metathesis groups concern total latent adduct; the E/Z-reactant group concerns the combined alkene isomers after fragmentation. Do not combine these endpoints across groups."
6
  },
7
  "measurement": {
8
  "name": "product 1H NMR yield",
9
  "unit": "%",
10
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
11
  },
12
  "measurements_by_group": {
13
  "g1": {
14
  "name": "product 1H NMR yield",
15
  "unit": "%",
16
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
17
  },
18
  "g2": {
19
  "name": "product 1H NMR yield",
20
  "unit": "%",
21
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
22
  },
23
  "g3": {
24
  "name": "product 1H NMR yield",
25
  "unit": "%",
26
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
27
  }
28
  },
29
  "measurement_rule": "Use measurements_by_group for each group. The measurement field is a default summary and does not override a group-specific endpoint.",
30
  "inheritance_rule": "common_conditions apply unless the same field is explicitly overridden by the item; item settings are literal experimental facts, not predicted outcomes.",
31
  "items": {
32
+ "exp_6c7c207a3e35": {
33
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
34
  "transfer_reagent": "1a",
35
+ "transfer_reagent_equiv": 3,
36
+ "catalyst": "HG-II as structures/metathesis_catalysts.png",
37
  "catalyst_mol_percent": 5,
38
  "solvent": "dichloromethane",
39
  "solvent_volume_mL": 1,
40
  "alkene_concentration_M": 0.1,
41
  "temperature_C": 40,
42
  "time_h": 8,
43
+ "procedure": "5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
44
+ "analysis": "Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
45
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
46
  },
47
+ "exp_777d484cc14b": {
48
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
49
  "transfer_reagent": "1a",
50
  "transfer_reagent_equiv": 2,
 
55
  "alkene_concentration_M": 0.1,
56
  "temperature_C": 40,
57
  "time_h": 8,
58
+ "procedure": "5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
59
+ "analysis": "Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
60
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
61
  },
62
+ "exp_7ee0f9400be1": {
63
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
64
  "transfer_reagent": "1a",
65
  "transfer_reagent_equiv": 1,
 
70
  "alkene_concentration_M": 0.1,
71
  "temperature_C": 40,
72
  "time_h": 8,
73
+ "procedure": "5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
74
+ "analysis": "Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
75
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
76
  },
77
+ "exp_b25afc3d08f5": {
78
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
79
  "transfer_reagent": "1a",
80
  "transfer_reagent_equiv": 4,
 
85
  "alkene_concentration_M": 0.1,
86
  "temperature_C": 40,
87
  "time_h": 8,
88
+ "procedure": "5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
89
+ "analysis": "Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
91
  },
92
+ "exp_ecccb58cf20b": {
93
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
94
+ "transfer_reagent": "1a",
95
+ "transfer_reagent_equiv": 4,
96
+ "catalyst": "G-II as structures/metathesis_catalysts.png",
97
  "catalyst_mol_percent": 5,
98
  "solvent": "dichloromethane",
99
  "solvent_volume_mL": 1,
100
  "alkene_concentration_M": 0.1,
101
  "temperature_C": 40,
102
  "time_h": 8,
103
+ "procedure": "5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
104
+ "analysis": "Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
105
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
106
  },
107
+ "exp_f6a85f2013bc": {
108
  "alkene": "4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",
109
+ "transfer_reagent": "1b",
110
  "transfer_reagent_equiv": 3,
111
  "catalyst": "HG-II as structures/metathesis_catalysts.png",
112
  "catalyst_mol_percent": 5,
113
+ "solvent": "toluene",
114
  "solvent_volume_mL": 1,
115
  "alkene_concentration_M": 0.1,
116
+ "temperature_C": 100,
117
  "time_h": 8,
118
+ "procedure": "5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.",
119
+ "analysis": "Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.",
120
  "product": "PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation."
121
  },
122
+ "exp_039643486694": {
123
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
124
+ "transfer_reagent": "1a",
125
+ "transfer_reagent_equiv": 10,
126
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
127
  "catalyst_mol_percent": 20,
128
  "solvent": "1,2-dichloroethane",
129
  "solvent_volume_mL": 0.5,
130
  "alkene_concentration_M": 0.1,
131
  "temperature_C": 100,
132
+ "time_h": 5,
133
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
134
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
135
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
136
  },
137
+ "exp_064708d49167": {
138
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
139
+ "transfer_reagent": "1b",
140
+ "transfer_reagent_equiv": 5,
141
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
142
  "catalyst_mol_percent": 20,
143
  "solvent": "1,2-dichloroethane",
144
  "solvent_volume_mL": 0.5,
145
  "alkene_concentration_M": 0.1,
146
  "temperature_C": 100,
147
+ "time_h": 16,
148
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
149
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
150
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
151
  },
152
+ "exp_121a21fd986b": {
153
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
154
  "transfer_reagent": "1b",
155
+ "transfer_reagent_equiv": 3,
156
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
157
  "catalyst_mol_percent": 20,
158
  "solvent": "1,2-dichloroethane",
 
160
  "alkene_concentration_M": 0.1,
161
  "temperature_C": 100,
162
  "time_h": 16,
163
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
164
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
165
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
166
  },
167
+ "exp_3c5d2113087d": {
168
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
169
  "transfer_reagent": "1a",
170
  "transfer_reagent_equiv": 10,
 
175
  "alkene_concentration_M": 0.1,
176
  "temperature_C": 100,
177
  "time_h": 5,
178
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
179
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
180
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
181
  },
182
+ "exp_5fe231ef098e": {
183
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
184
  "transfer_reagent": "1a",
185
  "transfer_reagent_equiv": 10,
 
190
  "alkene_concentration_M": 0.1,
191
  "temperature_C": 100,
192
  "time_h": 16,
193
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
194
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
195
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
196
  },
197
+ "exp_a3d319499eac": {
198
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
199
  "transfer_reagent": "1b",
200
+ "transfer_reagent_equiv": 2,
201
  "catalyst": "HG-II as structures/rotenone_catalysts.png",
202
  "catalyst_mol_percent": 20,
203
  "solvent": "1,2-dichloroethane",
 
205
  "alkene_concentration_M": 0.1,
206
  "temperature_C": 100,
207
  "time_h": 16,
208
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
209
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
210
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
211
  },
212
+ "exp_dddc25cdd5a3": {
213
  "alkene": "Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",
214
  "transfer_reagent": "1a",
215
  "transfer_reagent_equiv": 10,
 
220
  "alkene_concentration_M": 0.1,
221
  "temperature_C": 40,
222
  "time_h": 16,
223
+ "procedure": "10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.",
224
+ "analysis": "Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.",
225
  "product": "Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation."
226
  },
227
+ "exp_b167406216a1": {
228
+ "latent_precursor": "Z-17,fullmolecularstructure in structures/latent_Z.png,36.8mg0.100mmol",
229
  "citric_acid": "57.6mg0.300mmol3equiv",
230
  "fluoride": "Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv",
231
  "solvent": "0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",
232
  "substrate_concentration_M": 0.1,
233
  "temperature_C": 40,
234
  "time_h": 16,
235
+ "procedure": "10mLmicrowave vial with magnetic bar. Evacuate/refill N2 three cycles. Adddichloromethane, heatin oilbath. Thenaddcitricacid andTBAFtrihydrate dissolved in methanol through septum; heat and stir16h.",
236
+ "analysis": "Addmethylbenzoate12.6microlitre0.100mmol. Take crude aliquot for 1H NMR combined product yield; GC-MS separately determinesisomer ratio, but ratio is not the requested endpoint.",
237
  "product": "CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation."
238
  },
239
+ "exp_b81cc665557a": {
240
+ "latent_precursor": "E-17,fullmolecularstructure in structures/latent_E.png,36.8mg0.100mmol",
241
  "citric_acid": "57.6mg0.300mmol3equiv",
242
  "fluoride": "Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv",
243
  "solvent": "0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",
244
  "substrate_concentration_M": 0.1,
245
  "temperature_C": 40,
246
  "time_h": 16,
247
+ "procedure": "10mLmicrowave vial with magnetic bar. Evacuate/refill N2 three cycles. Adddichloromethane, heatin oilbath. Thenaddcitricacid andTBAFtrihydrate dissolved in methanol through septum; heat and stir16h.",
248
+ "analysis": "Addmethylbenzoate12.6microlitre0.100mmol. Take crude aliquot for 1H NMR combined product yield; GC-MS separately determinesisomer ratio, but ratio is not the requested endpoint.",
249
  "product": "CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation."
250
  }
251
  }
tasks/alkene-transfer-conditions-ranking/environment/public/targets.csv CHANGED
@@ -1,17 +1,16 @@
1
  group_id,item_id,settings_ref,alkene,transfer_reagent,transfer_reagent_equiv,catalyst,catalyst_mol_percent,solvent,solvent_volume_mL,alkene_concentration_M,temperature_C,time_h,procedure,analysis,product,latent_precursor,citric_acid,fluoride,substrate_concentration_M
2
- g1,exp_41bd98dba742,protocol.json#exp_41bd98dba742,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,4,G-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
3
- g1,exp_484d7efaebd5,protocol.json#exp_484d7efaebd5,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,2,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
4
- g1,exp_389c830f7e89,protocol.json#exp_389c830f7e89,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,1,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
5
- g1,exp_dd127eb2b1f5,protocol.json#exp_dd127eb2b1f5,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,4,G-I as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
6
- g1,exp_8545262bd121,protocol.json#exp_8545262bd121,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1b,3,HG-II as structures/metathesis_catalysts.png,5,toluene,1,0.1,100,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
7
- g1,exp_369ffbca2e28,protocol.json#exp_369ffbca2e28,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1b,3,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
8
- g1,exp_cac98fcd26c1,protocol.json#exp_cac98fcd26c1,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,3,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottomflask,magneticbar,refluxcondenser withTeflontapedjoint,suba-seal,N2line andoutletneedle toventethylene.Heat reagents in oilbath,add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
9
- g2,exp_066e33c896b9,protocol.json#exp_066e33c896b9,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1b,3,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
10
- g2,exp_fba92a14e6bb,protocol.json#exp_fba92a14e6bb,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,5,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
11
- g2,exp_9279aa7cd389,protocol.json#exp_9279aa7cd389,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1b,2,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
12
- g2,exp_8e5983854cd1,protocol.json#exp_8e5983854cd1,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,nitro-GrelaI2SiPr as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,5,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
13
- g2,exp_343b2ac1c9a7,protocol.json#exp_343b2ac1c9a7,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
14
- g2,exp_e01ea673fa5d,protocol.json#exp_e01ea673fa5d,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1b,5,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
15
- g2,exp_ea300526f709,protocol.json#exp_ea300526f709,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,HG-II as structures/rotenone_catalysts.png,20,dichloromethane,0.5,0.1,40,16,"10mLmicrowavevial withmagneticbar.Charge reagents andcatalyst,crimpseal,evacuate/refillN2threecycles,addsolventviasyringe.Heatstir,thenquenchcatalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo,measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
16
- g3,exp_529d2754aa6a,protocol.json#exp_529d2754aa6a,,,,,,"0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",,,40,16,"10mLmicrowavevial withmagneticbar.Evacuate/refillN2threecycles.Adddichloromethane,heatin oilbath.Thenaddcitricacid andTBAFtrihydrate dissolved inmethanol throughseptum;heatandstir16h.","Addmethylbenzoate12.6microlitre0.100mmol.Takecrudealiquot for1H NMR combinedproductyield;GC-MS separatelydeterminesisomerratio,butratiois notthe requestedendpoint.",CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation.,"E-17,fullmolecularstructure in structures/latent_E.png,36.8mg0.100mmol",57.6mg0.300mmol3equiv,Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv,0.1
17
- g3,exp_94c775c85236,protocol.json#exp_94c775c85236,,,,,,"0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",,,40,16,"10mLmicrowavevial withmagneticbar.Evacuate/refillN2threecycles.Adddichloromethane,heatin oilbath.Thenaddcitricacid andTBAFtrihydrate dissolved inmethanol throughseptum;heatandstir16h.","Addmethylbenzoate12.6microlitre0.100mmol.Takecrudealiquot for1H NMR combinedproductyield;GC-MS separatelydeterminesisomerratio,butratiois notthe requestedendpoint.",CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation.,"Z-17,fullmolecularstructure in structures/latent_Z.png,36.8mg0.100mmol",57.6mg0.300mmol3equiv,Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv,0.1
 
1
  group_id,item_id,settings_ref,alkene,transfer_reagent,transfer_reagent_equiv,catalyst,catalyst_mol_percent,solvent,solvent_volume_mL,alkene_concentration_M,temperature_C,time_h,procedure,analysis,product,latent_precursor,citric_acid,fluoride,substrate_concentration_M
2
+ g1,exp_6c7c207a3e35,protocol.json#exp_6c7c207a3e35,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,3,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
3
+ g1,exp_777d484cc14b,protocol.json#exp_777d484cc14b,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,2,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
4
+ g1,exp_7ee0f9400be1,protocol.json#exp_7ee0f9400be1,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,1,HG-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
5
+ g1,exp_b25afc3d08f5,protocol.json#exp_b25afc3d08f5,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,4,G-I as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
6
+ g1,exp_ecccb58cf20b,protocol.json#exp_ecccb58cf20b,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1a,4,G-II as structures/metathesis_catalysts.png,5,dichloromethane,1,0.1,40,8,"5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
7
+ g1,exp_f6a85f2013bc,protocol.json#exp_f6a85f2013bc,"4-phenyl-1-butene,C=CCCc1ccccc1,15microlitre0.100mmol1equiv",1b,3,HG-II as structures/metathesis_catalysts.png,5,toluene,1,0.1,100,8,"5mLround-bottom flask, magnetic bar, reflux condenser with Teflon taped joint, suba-seal,N2 line and outlet needle to vent ethylene. Heat reagents in oilbath, add0.25mLsolvent,0.5mLcatalyststock then0.25mLsolvent.","Evaporate solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate12.6microlitre0.100mmol.","PhCH2CH2CH=CHCH2S(=O)2CH2OTBS,totalgeometricisomers,beforefragmentation.",,,,
8
+ g2,exp_039643486694,protocol.json#exp_039643486694,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,5,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
9
+ g2,exp_064708d49167,protocol.json#exp_064708d49167,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1b,5,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
10
+ g2,exp_121a21fd986b,protocol.json#exp_121a21fd986b,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1b,3,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
11
+ g2,exp_3c5d2113087d,protocol.json#exp_3c5d2113087d,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,nitro-GrelaI2SiPr as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,5,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
12
+ g2,exp_5fe231ef098e,protocol.json#exp_5fe231ef098e,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
13
+ g2,exp_a3d319499eac,protocol.json#exp_a3d319499eac,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1b,2,HG-II as structures/rotenone_catalysts.png,20,"1,2-dichloroethane",0.5,0.1,100,16,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
14
+ g2,exp_dddc25cdd5a3,protocol.json#exp_dddc25cdd5a3,"Rotenone,stereostructure shown in structures/rotenone_reaction.png,19.7mg0.050mmol1equiv",1a,10,HG-II as structures/rotenone_catalysts.png,20,dichloromethane,0.5,0.1,40,16,"10mLmicrowave vial with magnetic bar. Charge reagents and catalyst, crimpseal,evacuate/refill N2 three cycles, add solvent via syringe. Heat and stir, then quench catalyst with~0.1mLbutylvinylether.","Remove solvent in vacuo, measure1H NMR adduct yield againstmethylbenzoate6.3microlitre0.050mmol.","Rotenone cross-metathesis product in structures/rotenone_reaction.png,totalgeometricisomers,beforefragmentation.",,,,
15
+ g3,exp_b167406216a1,protocol.json#exp_b167406216a1,,,,,,"0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",,,40,16,"10mLmicrowave vial with magnetic bar. Evacuate/refill N2 three cycles. Adddichloromethane, heatin oilbath. Thenaddcitricacid andTBAFtrihydrate dissolved in methanol through septum; heat and stir16h.","Addmethylbenzoate12.6microlitre0.100mmol. Take crude aliquot for 1H NMR combined product yield; GC-MS separately determinesisomer ratio, but ratio is not the requested endpoint.",CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation.,"Z-17,fullmolecularstructure in structures/latent_Z.png,36.8mg0.100mmol",57.6mg0.300mmol3equiv,Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv,0.1
16
+ g3,exp_b81cc665557a,protocol.json#exp_b81cc665557a,,,,,,"0.5mLdichloromethane plus0.5mLmethanol,total1.0mL",,,40,16,"10mLmicrowave vial with magnetic bar. Evacuate/refill N2 three cycles. Adddichloromethane, heatin oilbath. Thenaddcitricacid andTBAFtrihydrate dissolved in methanol through septum; heat and stir16h.","Addmethylbenzoate12.6microlitre0.100mmol. Take crude aliquot for 1H NMR combined product yield; GC-MS separately determinesisomer ratio, but ratio is not the requested endpoint.",CombinedE+Z1-phenylhex-4-ene(C6H5CH2CH2CH2CH=CHCH3) afterfragmentation.,"E-17,fullmolecularstructure in structures/latent_E.png,36.8mg0.100mmol",57.6mg0.300mmol3equiv,Tetrabutylammoniumfluoride trihydrate39.4mg0.125mmol1.25equiv,0.1
 
tasks/alkene-transfer-conditions-ranking/instruction.md CHANGED
@@ -26,13 +26,13 @@ Combine each listed condition with the common procedure. Rank the specified meas
26
 
27
  Outcome definitions by group:
28
 
29
- - `g1` — product 1H NMR yield (%): Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard.
30
- - `g2` — product 1H NMR yield (%): Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard.
31
- - `g3` — product 1H NMR yield (%): Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard.
32
 
33
  ## Comparison groups
34
 
35
- - `g1`: 4-Phenyl-1-butene cross-metathesis adduct yield; 7 items; weight 0.3333333333333333; ascending product 1H NMR yield (%).
36
  - `g2`: Rotenone cross-metathesis adduct yield; 7 items; weight 0.3333333333333333; ascending product 1H NMR yield (%).
37
  - `g3`: Combined alkene yield from E/Z latent precursors; 2 items; weight 0.3333333333333333; ascending product 1H NMR yield (%).
38
 
@@ -55,127 +55,110 @@ You have 8 CPU cores, 16,384 MiB system RAM and one NVIDIA H20 GPU, shared by al
55
 
56
  ## Submission
57
 
58
- Write UTF-8 JSON to `/logs/artifacts/rankings.json` (at most 1 MiB, unique object keys). Include the exact task identity and exactly one entry per group; each order must contain every assigned item once, from lowest to highest. The following is a format example in public display order, not a predicted answer:
59
 
60
  ```json
61
  {
62
  "task_id": "alkene-transfer-conditions-ranking",
63
- "task_version": "1.0.0",
64
  "rankings": [
65
  {
66
  "group_id": "g1",
67
  "order": [
68
- "exp_41bd98dba742",
69
- "exp_484d7efaebd5",
70
- "exp_389c830f7e89",
71
- "exp_dd127eb2b1f5",
72
- "exp_8545262bd121",
73
- "exp_369ffbca2e28",
74
- "exp_cac98fcd26c1"
75
  ]
76
  },
77
  {
78
  "group_id": "g2",
79
  "order": [
80
- "exp_066e33c896b9",
81
- "exp_fba92a14e6bb",
82
- "exp_9279aa7cd389",
83
- "exp_8e5983854cd1",
84
- "exp_343b2ac1c9a7",
85
- "exp_e01ea673fa5d",
86
- "exp_ea300526f709"
87
  ]
88
  },
89
  {
90
  "group_id": "g3",
91
  "order": [
92
- "exp_529d2754aa6a",
93
- "exp_94c775c85236"
94
  ]
95
  }
96
  ]
97
  }
98
  ```
99
 
100
- Write an English account of scientific reasoning, calculations, assumptions and uncertainty to `/logs/artifacts/report.md`; source records to `/logs/artifacts/sources.json`; supporting scripts and calculations to `/logs/artifacts/analysis/`. The evaluator uses the last submission received before the deadline. Empty or missing submissions receive core -1 and auxiliary 0.
101
 
102
- ## Ranking scores
103
 
104
- The primary score is standard Spearman rank correlation, from -1 to 1, computed separately in each group and combined with fixed group weights. Equal measured values receive average true ranks; predictions remain strict permutations. Correct/reverse rankings score 1/-1 when true values have no ties; with true ties the best strict ranking can score less than 1. Random complete rankings have expected correlation 0.
105
 
106
- The auxiliary score counts each decidable pair equally within its group, combines group accuracies using the same weights, then applies one whole-task random 95th-percentile threshold. Known uncertainty uses the declared difference-error rule; unknown uncertainty uses reported values with epsilon and does not exclude unequal results. The threshold is hidden. Core and auxiliary scores are recorded separately; neither meaning of zero proves the absence of predictive ability. Detailed algorithms and all format penalties are in the following authoritative public contract, from which the companion files are generated.
107
-
108
- <!-- public-contract -->
109
  ```json
110
  {
111
  "task_id": "alkene-transfer-conditions-ranking",
112
- "task_version": "1.0.0",
113
  "groups": [
114
  {
115
  "group_id": "g1",
116
  "label": "4-Phenyl-1-butene cross-metathesis adduct yield",
117
  "item_ids": [
118
- "exp_41bd98dba742",
119
- "exp_484d7efaebd5",
120
- "exp_389c830f7e89",
121
- "exp_dd127eb2b1f5",
122
- "exp_8545262bd121",
123
- "exp_369ffbca2e28",
124
- "exp_cac98fcd26c1"
125
  ],
126
  "direction": "ascending",
127
  "group_weight": 0.3333333333333333,
128
- "k": 2,
129
- "epsilon": 0,
130
- "uncertainty_model": "optional_reported_uncertainty",
131
- "difference_standard_error": "sem_sum_upper_bound",
132
  "endpoint": {
133
  "name": "product 1H NMR yield",
134
  "unit": "%",
135
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
136
  }
137
  },
138
  {
139
  "group_id": "g2",
140
  "label": "Rotenone cross-metathesis adduct yield",
141
  "item_ids": [
142
- "exp_066e33c896b9",
143
- "exp_fba92a14e6bb",
144
- "exp_9279aa7cd389",
145
- "exp_8e5983854cd1",
146
- "exp_343b2ac1c9a7",
147
- "exp_e01ea673fa5d",
148
- "exp_ea300526f709"
149
  ],
150
  "direction": "ascending",
151
  "group_weight": 0.3333333333333333,
152
- "k": 2,
153
- "epsilon": 0,
154
- "uncertainty_model": "optional_reported_uncertainty",
155
- "difference_standard_error": "sem_sum_upper_bound",
156
  "endpoint": {
157
  "name": "product 1H NMR yield",
158
  "unit": "%",
159
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
160
  }
161
  },
162
  {
163
  "group_id": "g3",
164
  "label": "Combined alkene yield from E/Z latent precursors",
165
  "item_ids": [
166
- "exp_529d2754aa6a",
167
- "exp_94c775c85236"
168
  ],
169
  "direction": "ascending",
170
  "group_weight": 0.3333333333333333,
171
- "k": 2,
172
- "epsilon": 0,
173
- "uncertainty_model": "optional_reported_uncertainty",
174
- "difference_standard_error": "sem_sum_upper_bound",
175
  "endpoint": {
176
  "name": "product 1H NMR yield",
177
  "unit": "%",
178
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
179
  }
180
  }
181
  ],
@@ -188,18 +171,9 @@ The auxiliary score counts each decidable pair equally within its group, combine
188
  "spearman_tie_method": "average",
189
  "spearman_undefined_policy": "task_error",
190
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
191
- "secondary_scoring_method": "equal_pair_inversion_random95",
192
- "ranking_reward_range": [
193
- 0,
194
- 1
195
- ],
196
- "random_quantile": 0.95,
197
- "unknown_uncertainty_policy": "use_reported_value",
198
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
199
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
200
  "format_errors": {
201
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
202
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
203
  },
204
  "max_submission_bytes": 1048576,
205
  "minimum_conditions_per_group": 2,
@@ -216,7 +190,16 @@ The auxiliary score counts each decidable pair equally within its group, combine
216
  "report_path": "/logs/artifacts/report.md",
217
  "sources_path": "/logs/artifacts/sources.json",
218
  "analysis_path": "/logs/artifacts/analysis/",
219
- "forbidden_result_scope": "Measured outcomes, yields, rankings, equivalent endpoints and result-fitted predictors for alkene transfer-reagent and catalyst conditions in the stated material and reaction system are prohibited."
 
 
 
 
 
 
 
 
 
220
  }
221
  ```
222
  <!-- /public-contract -->
 
26
 
27
  Outcome definitions by group:
28
 
29
+ - `g1` — product 1H NMR yield (%): Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard.
30
+ - `g2` — product 1H NMR yield (%): Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard.
31
+ - `g3` — product 1H NMR yield (%): Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard.
32
 
33
  ## Comparison groups
34
 
35
+ - `g1`: 4-Phenyl-1-butene cross-metathesis adduct yield; 6 items; weight 0.3333333333333333; ascending product 1H NMR yield (%).
36
  - `g2`: Rotenone cross-metathesis adduct yield; 7 items; weight 0.3333333333333333; ascending product 1H NMR yield (%).
37
  - `g3`: Combined alkene yield from E/Z latent precursors; 2 items; weight 0.3333333333333333; ascending product 1H NMR yield (%).
38
 
 
55
 
56
  ## Submission
57
 
58
+ Write UTF-8 JSON to `/logs/artifacts/rankings.json` (at most 1 MiB, unique object keys). Include the exact task identity and exactly one entry per group; each order must contain every assigned item once, from lowest to highest. The example illustrates the submission format. Submit your predicted order for each group.
59
 
60
  ```json
61
  {
62
  "task_id": "alkene-transfer-conditions-ranking",
63
+ "task_version": "1.1.0",
64
  "rankings": [
65
  {
66
  "group_id": "g1",
67
  "order": [
68
+ "exp_6c7c207a3e35",
69
+ "exp_777d484cc14b",
70
+ "exp_7ee0f9400be1",
71
+ "exp_b25afc3d08f5",
72
+ "exp_ecccb58cf20b",
73
+ "exp_f6a85f2013bc"
 
74
  ]
75
  },
76
  {
77
  "group_id": "g2",
78
  "order": [
79
+ "exp_039643486694",
80
+ "exp_064708d49167",
81
+ "exp_121a21fd986b",
82
+ "exp_3c5d2113087d",
83
+ "exp_5fe231ef098e",
84
+ "exp_a3d319499eac",
85
+ "exp_dddc25cdd5a3"
86
  ]
87
  },
88
  {
89
  "group_id": "g3",
90
  "order": [
91
+ "exp_b167406216a1",
92
+ "exp_b81cc665557a"
93
  ]
94
  }
95
  ]
96
  }
97
  ```
98
 
99
+ Write an English account of scientific reasoning, calculations, assumptions and uncertainty to `/logs/artifacts/report.md`; source records to `/logs/artifacts/sources.json`; supporting scripts and calculations to `/logs/artifacts/analysis/`. Only rankings.json contributes to the Spearman reward. The verifier records the presence of the supporting artifacts for review. For a stop under the information policy, put `SPU_POLICY_STOP` on the first line of `report.md` and record the triggering step, source and reason below it. Withdraw any provisional ranking. The evaluator uses the last submission received before the deadline. Empty or missing submissions receive -1.
100
 
101
+ ## Ranking score
102
 
103
+ The task score is the weighted mean of the group Spearman rank correlations and ranges from -1 to 1. Equal measurements share their average rank. Submit a strict ordering for each group. When measurements are distinct, the measured order scores 1 and its reverse scores -1. Tied measurements can place the highest attainable correlation below 1. A uniformly random ordering has expected correlation 0.
104
 
 
 
 
105
  ```json
106
  {
107
  "task_id": "alkene-transfer-conditions-ranking",
108
+ "task_version": "1.1.0",
109
  "groups": [
110
  {
111
  "group_id": "g1",
112
  "label": "4-Phenyl-1-butene cross-metathesis adduct yield",
113
  "item_ids": [
114
+ "exp_6c7c207a3e35",
115
+ "exp_777d484cc14b",
116
+ "exp_7ee0f9400be1",
117
+ "exp_b25afc3d08f5",
118
+ "exp_ecccb58cf20b",
119
+ "exp_f6a85f2013bc"
 
120
  ],
121
  "direction": "ascending",
122
  "group_weight": 0.3333333333333333,
 
 
 
 
123
  "endpoint": {
124
  "name": "product 1H NMR yield",
125
  "unit": "%",
126
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
127
  }
128
  },
129
  {
130
  "group_id": "g2",
131
  "label": "Rotenone cross-metathesis adduct yield",
132
  "item_ids": [
133
+ "exp_039643486694",
134
+ "exp_064708d49167",
135
+ "exp_121a21fd986b",
136
+ "exp_3c5d2113087d",
137
+ "exp_5fe231ef098e",
138
+ "exp_a3d319499eac",
139
+ "exp_dddc25cdd5a3"
140
  ],
141
  "direction": "ascending",
142
  "group_weight": 0.3333333333333333,
 
 
 
 
143
  "endpoint": {
144
  "name": "product 1H NMR yield",
145
  "unit": "%",
146
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
147
  }
148
  },
149
  {
150
  "group_id": "g3",
151
  "label": "Combined alkene yield from E/Z latent precursors",
152
  "item_ids": [
153
+ "exp_b167406216a1",
154
+ "exp_b81cc665557a"
155
  ],
156
  "direction": "ascending",
157
  "group_weight": 0.3333333333333333,
 
 
 
 
158
  "endpoint": {
159
  "name": "product 1H NMR yield",
160
  "unit": "%",
161
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
162
  }
163
  }
164
  ],
 
171
  "spearman_tie_method": "average",
172
  "spearman_undefined_policy": "task_error",
173
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
174
  "format_errors": {
175
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
176
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
177
  },
178
  "max_submission_bytes": 1048576,
179
  "minimum_conditions_per_group": 2,
 
190
  "report_path": "/logs/artifacts/report.md",
191
  "sources_path": "/logs/artifacts/sources.json",
192
  "analysis_path": "/logs/artifacts/analysis/",
193
+ "forbidden_result_scope": "Measured outcomes, yields, rankings, equivalent endpoints and result-fitted predictors for alkene transfer-reagent and catalyst conditions in the stated material and reaction system are prohibited.",
194
+ "supporting_artifact_review": {
195
+ "policy_stop_marker": "SPU_POLICY_STOP",
196
+ "required_files": [
197
+ "report.md",
198
+ "sources.json"
199
+ ],
200
+ "required_directory": "analysis",
201
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
202
+ }
203
  }
204
  ```
205
  <!-- /public-contract -->
tasks/alkene-transfer-conditions-ranking/solution/solve.sh CHANGED
@@ -1,42 +1,41 @@
1
  #!/bin/sh
2
  set -eu
3
  mkdir -p /logs/artifacts
4
- cat > /logs/artifacts/rankings.json <<'PROJECTAA_ORACLE'
5
  {
6
  "task_id": "alkene-transfer-conditions-ranking",
7
- "task_version": "1.0.0",
8
  "rankings": [
9
  {
10
  "group_id": "g1",
11
  "order": [
12
- "exp_dd127eb2b1f5",
13
- "exp_369ffbca2e28",
14
- "exp_8545262bd121",
15
- "exp_389c830f7e89",
16
- "exp_484d7efaebd5",
17
- "exp_cac98fcd26c1",
18
- "exp_41bd98dba742"
19
  ]
20
  },
21
  {
22
  "group_id": "g2",
23
  "order": [
24
- "exp_8e5983854cd1",
25
- "exp_343b2ac1c9a7",
26
- "exp_ea300526f709",
27
- "exp_fba92a14e6bb",
28
- "exp_9279aa7cd389",
29
- "exp_e01ea673fa5d",
30
- "exp_066e33c896b9"
31
  ]
32
  },
33
  {
34
  "group_id": "g3",
35
  "order": [
36
- "exp_94c775c85236",
37
- "exp_529d2754aa6a"
38
  ]
39
  }
40
  ]
41
  }
42
- PROJECTAA_ORACLE
 
1
  #!/bin/sh
2
  set -eu
3
  mkdir -p /logs/artifacts
4
+ cat > /logs/artifacts/rankings.json <<'EOF'
5
  {
6
  "task_id": "alkene-transfer-conditions-ranking",
7
+ "task_version": "1.1.0",
8
  "rankings": [
9
  {
10
  "group_id": "g1",
11
  "order": [
12
+ "exp_b25afc3d08f5",
13
+ "exp_f6a85f2013bc",
14
+ "exp_7ee0f9400be1",
15
+ "exp_777d484cc14b",
16
+ "exp_6c7c207a3e35",
17
+ "exp_ecccb58cf20b"
 
18
  ]
19
  },
20
  {
21
  "group_id": "g2",
22
  "order": [
23
+ "exp_3c5d2113087d",
24
+ "exp_5fe231ef098e",
25
+ "exp_dddc25cdd5a3",
26
+ "exp_039643486694",
27
+ "exp_a3d319499eac",
28
+ "exp_064708d49167",
29
+ "exp_121a21fd986b"
30
  ]
31
  },
32
  {
33
  "group_id": "g3",
34
  "order": [
35
+ "exp_b167406216a1",
36
+ "exp_b81cc665557a"
37
  ]
38
  }
39
  ]
40
  }
41
+ EOF
tasks/alkene-transfer-conditions-ranking/task.toml CHANGED
@@ -1,9 +1,7 @@
1
- schema_version = "1.2"
2
  [metadata]
3
  task_id = "alkene-transfer-conditions-ranking"
4
- task_version = "1.0.0"
5
- pipeline_version = "19"
6
- scientist_prompt_version = "projectaa-scientist-en-v7-ranking"
7
  [agent]
8
  user = "solver"
9
  timeout_sec = 28800.0
@@ -14,7 +12,7 @@ memory_mb = 16384
14
  storage_mb = 20480
15
  gpus = 1
16
  gpu_types = ["H20"]
17
- allow_internet = true
18
  [verifier]
19
  timeout_sec = 120.0
20
  environment_mode = "separate"
@@ -24,4 +22,4 @@ cpus = 1
24
  memory_mb = 1024
25
  storage_mb = 2048
26
  gpus = 0
27
- allow_internet = false
 
1
+ schema_version = "1.4"
2
  [metadata]
3
  task_id = "alkene-transfer-conditions-ranking"
4
+ task_version = "1.1.0"
 
 
5
  [agent]
6
  user = "solver"
7
  timeout_sec = 28800.0
 
12
  storage_mb = 20480
13
  gpus = 1
14
  gpu_types = ["H20"]
15
+ network_mode = "public"
16
  [verifier]
17
  timeout_sec = 120.0
18
  environment_mode = "separate"
 
22
  memory_mb = 1024
23
  storage_mb = 2048
24
  gpus = 0
25
+ network_mode = "no-network"
tasks/alkene-transfer-conditions-ranking/tests/answers.json CHANGED
@@ -1,644 +1,126 @@
1
  {
2
  "task_id": "alkene-transfer-conditions-ranking",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "g1",
7
  "items": [
8
  {
9
- "item_id": "exp_41bd98dba742",
10
- "mean": 97,
11
  "n": null,
12
  "uncertainty_status": "unknown",
13
- "uncertainty_scale": null,
14
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
15
- "quality_status": "retained_with_flag",
16
- "warnings": [
17
- "Independent-mean uncertainty is unknown."
18
- ],
19
- "source_location": "Original-SI.pdf physical p6 TableS3 entry2,catalyst structures;p5section3.1method",
20
- "aggregation": "Published endpoint value without reweighting or selection of readings."
21
  },
22
  {
23
- "item_id": "exp_484d7efaebd5",
24
  "mean": 89,
25
  "n": null,
26
  "uncertainty_status": "unknown",
27
- "uncertainty_scale": null,
28
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
29
- "quality_status": "retained_with_flag",
30
- "warnings": [
31
- "Independent-mean uncertainty is unknown."
32
- ],
33
- "source_location": "Original-SI.pdf physical p5 TableS1 entry2,scheme andsection3.1method",
34
- "aggregation": "Published endpoint value without reweighting or selection of readings."
35
  },
36
  {
37
- "item_id": "exp_389c830f7e89",
38
  "mean": 66,
39
  "n": null,
40
  "uncertainty_status": "unknown",
41
- "uncertainty_scale": null,
42
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
43
- "quality_status": "retained_with_flag",
44
- "warnings": [
45
- "Independent-mean uncertainty is unknown."
46
- ],
47
- "source_location": "Original-SI.pdf physical p5 TableS1 entry1,scheme andsection3.1method",
48
- "aggregation": "Published endpoint value without reweighting or selection of readings."
49
  },
50
  {
51
- "item_id": "exp_dd127eb2b1f5",
52
  "mean": 11,
53
  "n": null,
54
  "uncertainty_status": "unknown",
55
- "uncertainty_scale": null,
56
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
57
- "quality_status": "retained_with_flag",
58
- "warnings": [
59
- "Independent-mean uncertainty is unknown."
60
- ],
61
- "source_location": "Original-SI.pdf physical p6 TableS3 entry3,catalyst structures;p5section3.1method",
62
- "aggregation": "Published endpoint value without reweighting or selection of readings."
63
  },
64
  {
65
- "item_id": "exp_8545262bd121",
66
- "mean": 53,
67
  "n": null,
68
  "uncertainty_status": "unknown",
69
- "uncertainty_scale": null,
70
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
71
- "quality_status": "retained_with_flag",
72
- "warnings": [
73
- "Independent-mean uncertainty is unknown."
74
- ],
75
- "source_location": "Original-SI.pdf physical p6 TableS4 entry3,scheme;p5section3.1method",
76
- "aggregation": "Published endpoint value without reweighting or selection of readings."
77
  },
78
  {
79
- "item_id": "exp_369ffbca2e28",
80
  "mean": 53,
81
  "n": null,
82
  "uncertainty_status": "unknown",
83
- "uncertainty_scale": null,
84
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
85
- "quality_status": "retained_with_flag",
86
- "warnings": [
87
- "Independent-mean uncertainty is unknown."
88
- ],
89
- "source_location": "Original-SI.pdf physical p6 TableS4 entry2,scheme;p5section3.1method",
90
- "aggregation": "Published endpoint value without reweighting or selection of readings."
91
- },
92
- {
93
- "item_id": "exp_cac98fcd26c1",
94
- "mean": 95,
95
- "n": null,
96
- "uncertainty_status": "unknown",
97
- "uncertainty_scale": null,
98
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
99
- "quality_status": "retained_with_flag",
100
- "warnings": [
101
- "Independent-mean uncertainty is unknown."
102
- ],
103
- "source_location": "Original-SI.pdf physical p5 TableS1 entry3,scheme andsection3.1method",
104
- "aggregation": "Published endpoint value without reweighting or selection of readings."
105
  }
106
- ],
107
- "pairs": [
108
- {
109
- "item_i": "exp_41bd98dba742",
110
- "item_j": "exp_484d7efaebd5",
111
- "mean_difference": 8,
112
- "difference_se": null,
113
- "uncertainty_status": "unknown",
114
- "comparison_basis": "reported_value",
115
- "comparable": 1
116
- },
117
- {
118
- "item_i": "exp_41bd98dba742",
119
- "item_j": "exp_389c830f7e89",
120
- "mean_difference": 31,
121
- "difference_se": null,
122
- "uncertainty_status": "unknown",
123
- "comparison_basis": "reported_value",
124
- "comparable": 1
125
- },
126
- {
127
- "item_i": "exp_41bd98dba742",
128
- "item_j": "exp_dd127eb2b1f5",
129
- "mean_difference": 86,
130
- "difference_se": null,
131
- "uncertainty_status": "unknown",
132
- "comparison_basis": "reported_value",
133
- "comparable": 1
134
- },
135
- {
136
- "item_i": "exp_41bd98dba742",
137
- "item_j": "exp_8545262bd121",
138
- "mean_difference": 44,
139
- "difference_se": null,
140
- "uncertainty_status": "unknown",
141
- "comparison_basis": "reported_value",
142
- "comparable": 1
143
- },
144
- {
145
- "item_i": "exp_41bd98dba742",
146
- "item_j": "exp_369ffbca2e28",
147
- "mean_difference": 44,
148
- "difference_se": null,
149
- "uncertainty_status": "unknown",
150
- "comparison_basis": "reported_value",
151
- "comparable": 1
152
- },
153
- {
154
- "item_i": "exp_41bd98dba742",
155
- "item_j": "exp_cac98fcd26c1",
156
- "mean_difference": 2,
157
- "difference_se": null,
158
- "uncertainty_status": "unknown",
159
- "comparison_basis": "reported_value",
160
- "comparable": 1
161
- },
162
- {
163
- "item_i": "exp_484d7efaebd5",
164
- "item_j": "exp_389c830f7e89",
165
- "mean_difference": 23,
166
- "difference_se": null,
167
- "uncertainty_status": "unknown",
168
- "comparison_basis": "reported_value",
169
- "comparable": 1
170
- },
171
- {
172
- "item_i": "exp_484d7efaebd5",
173
- "item_j": "exp_dd127eb2b1f5",
174
- "mean_difference": 78,
175
- "difference_se": null,
176
- "uncertainty_status": "unknown",
177
- "comparison_basis": "reported_value",
178
- "comparable": 1
179
- },
180
- {
181
- "item_i": "exp_484d7efaebd5",
182
- "item_j": "exp_8545262bd121",
183
- "mean_difference": 36,
184
- "difference_se": null,
185
- "uncertainty_status": "unknown",
186
- "comparison_basis": "reported_value",
187
- "comparable": 1
188
- },
189
- {
190
- "item_i": "exp_484d7efaebd5",
191
- "item_j": "exp_369ffbca2e28",
192
- "mean_difference": 36,
193
- "difference_se": null,
194
- "uncertainty_status": "unknown",
195
- "comparison_basis": "reported_value",
196
- "comparable": 1
197
- },
198
- {
199
- "item_i": "exp_484d7efaebd5",
200
- "item_j": "exp_cac98fcd26c1",
201
- "mean_difference": -6,
202
- "difference_se": null,
203
- "uncertainty_status": "unknown",
204
- "comparison_basis": "reported_value",
205
- "comparable": 1
206
- },
207
- {
208
- "item_i": "exp_389c830f7e89",
209
- "item_j": "exp_dd127eb2b1f5",
210
- "mean_difference": 55,
211
- "difference_se": null,
212
- "uncertainty_status": "unknown",
213
- "comparison_basis": "reported_value",
214
- "comparable": 1
215
- },
216
- {
217
- "item_i": "exp_389c830f7e89",
218
- "item_j": "exp_8545262bd121",
219
- "mean_difference": 13,
220
- "difference_se": null,
221
- "uncertainty_status": "unknown",
222
- "comparison_basis": "reported_value",
223
- "comparable": 1
224
- },
225
- {
226
- "item_i": "exp_389c830f7e89",
227
- "item_j": "exp_369ffbca2e28",
228
- "mean_difference": 13,
229
- "difference_se": null,
230
- "uncertainty_status": "unknown",
231
- "comparison_basis": "reported_value",
232
- "comparable": 1
233
- },
234
- {
235
- "item_i": "exp_389c830f7e89",
236
- "item_j": "exp_cac98fcd26c1",
237
- "mean_difference": -29,
238
- "difference_se": null,
239
- "uncertainty_status": "unknown",
240
- "comparison_basis": "reported_value",
241
- "comparable": 1
242
- },
243
- {
244
- "item_i": "exp_dd127eb2b1f5",
245
- "item_j": "exp_8545262bd121",
246
- "mean_difference": -42,
247
- "difference_se": null,
248
- "uncertainty_status": "unknown",
249
- "comparison_basis": "reported_value",
250
- "comparable": 1
251
- },
252
- {
253
- "item_i": "exp_dd127eb2b1f5",
254
- "item_j": "exp_369ffbca2e28",
255
- "mean_difference": -42,
256
- "difference_se": null,
257
- "uncertainty_status": "unknown",
258
- "comparison_basis": "reported_value",
259
- "comparable": 1
260
- },
261
- {
262
- "item_i": "exp_dd127eb2b1f5",
263
- "item_j": "exp_cac98fcd26c1",
264
- "mean_difference": -84,
265
- "difference_se": null,
266
- "uncertainty_status": "unknown",
267
- "comparison_basis": "reported_value",
268
- "comparable": 1
269
- },
270
- {
271
- "item_i": "exp_8545262bd121",
272
- "item_j": "exp_369ffbca2e28",
273
- "mean_difference": 0,
274
- "difference_se": null,
275
- "uncertainty_status": "unknown",
276
- "comparison_basis": "reported_value",
277
- "comparable": 0
278
- },
279
- {
280
- "item_i": "exp_8545262bd121",
281
- "item_j": "exp_cac98fcd26c1",
282
- "mean_difference": -42,
283
- "difference_se": null,
284
- "uncertainty_status": "unknown",
285
- "comparison_basis": "reported_value",
286
- "comparable": 1
287
- },
288
- {
289
- "item_i": "exp_369ffbca2e28",
290
- "item_j": "exp_cac98fcd26c1",
291
- "mean_difference": -42,
292
- "difference_se": null,
293
- "uncertainty_status": "unknown",
294
- "comparison_basis": "reported_value",
295
- "comparable": 1
296
- }
297
- ],
298
- "comparable_pairs": 20,
299
- "quality_status": "passed"
300
  },
301
  {
302
  "group_id": "g2",
303
  "items": [
304
  {
305
- "item_id": "exp_066e33c896b9",
306
- "mean": 94,
307
  "n": null,
308
  "uncertainty_status": "unknown",
309
- "uncertainty_scale": null,
310
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
311
- "quality_status": "retained_with_flag",
312
- "warnings": [
313
- "Independent-mean uncertainty is unknown."
314
- ],
315
- "source_location": "Original-SI.pdf physical p9 TableS6 entry7,reaction/catalyststructures andsection3.3procedure",
316
- "aggregation": "Published endpoint value without reweighting or selection of readings."
317
  },
318
  {
319
- "item_id": "exp_fba92a14e6bb",
320
- "mean": 48,
321
  "n": null,
322
  "uncertainty_status": "unknown",
323
- "uncertainty_scale": null,
324
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
325
- "quality_status": "retained_with_flag",
326
- "warnings": [
327
- "Independent-mean uncertainty is unknown."
328
- ],
329
- "source_location": "Original-SI.pdf physical p9 TableS6 entry2,reaction/catalyststructures andsection3.3procedure",
330
- "aggregation": "Published endpoint value without reweighting or selection of readings."
331
  },
332
  {
333
- "item_id": "exp_9279aa7cd389",
334
- "mean": 76,
335
  "n": null,
336
  "uncertainty_status": "unknown",
337
- "uncertainty_scale": null,
338
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
339
- "quality_status": "retained_with_flag",
340
- "warnings": [
341
- "Independent-mean uncertainty is unknown."
342
- ],
343
- "source_location": "Original-SI.pdf physical p9 TableS6 entry6,reaction/catalyststructures andsection3.3procedure",
344
- "aggregation": "Published endpoint value without reweighting or selection of readings."
345
  },
346
  {
347
- "item_id": "exp_8e5983854cd1",
348
  "mean": 5,
349
  "n": null,
350
  "uncertainty_status": "unknown",
351
- "uncertainty_scale": null,
352
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
353
- "quality_status": "retained_with_flag",
354
- "warnings": [
355
- "Independent-mean uncertainty is unknown."
356
- ],
357
- "source_location": "Original-SI.pdf physical p9 TableS6 entry5,reaction/catalyststructures andsection3.3procedure",
358
- "aggregation": "Published endpoint value without reweighting or selection of readings."
359
  },
360
  {
361
- "item_id": "exp_343b2ac1c9a7",
362
  "mean": 38,
363
  "n": null,
364
  "uncertainty_status": "unknown",
365
- "uncertainty_scale": null,
366
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
367
- "quality_status": "retained_with_flag",
368
- "warnings": [
369
- "Independent-mean uncertainty is unknown."
370
- ],
371
- "source_location": "Original-SI.pdf physical p9 TableS6 entry3,reaction/catalyststructures andsection3.3procedure",
372
- "aggregation": "Published endpoint value without reweighting or selection of readings."
373
  },
374
  {
375
- "item_id": "exp_e01ea673fa5d",
376
- "mean": 90,
377
  "n": null,
378
  "uncertainty_status": "unknown",
379
- "uncertainty_scale": null,
380
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
381
- "quality_status": "retained_with_flag",
382
- "warnings": [
383
- "Independent-mean uncertainty is unknown."
384
- ],
385
- "source_location": "Original-SI.pdf physical p9 TableS6 entry8,reaction/catalyststructures andsection3.3procedure",
386
- "aggregation": "Published endpoint value without reweighting or selection of readings."
387
  },
388
  {
389
- "item_id": "exp_ea300526f709",
390
  "mean": 44,
391
  "n": null,
392
  "uncertainty_status": "unknown",
393
- "uncertainty_scale": null,
394
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
395
- "quality_status": "retained_with_flag",
396
- "warnings": [
397
- "Independent-mean uncertainty is unknown."
398
- ],
399
- "source_location": "Original-SI.pdf physical p9 TableS6 entry1,reaction/catalyststructures andsection3.3procedure",
400
- "aggregation": "Published endpoint value without reweighting or selection of readings."
401
  }
402
- ],
403
- "pairs": [
404
- {
405
- "item_i": "exp_066e33c896b9",
406
- "item_j": "exp_fba92a14e6bb",
407
- "mean_difference": 46,
408
- "difference_se": null,
409
- "uncertainty_status": "unknown",
410
- "comparison_basis": "reported_value",
411
- "comparable": 1
412
- },
413
- {
414
- "item_i": "exp_066e33c896b9",
415
- "item_j": "exp_9279aa7cd389",
416
- "mean_difference": 18,
417
- "difference_se": null,
418
- "uncertainty_status": "unknown",
419
- "comparison_basis": "reported_value",
420
- "comparable": 1
421
- },
422
- {
423
- "item_i": "exp_066e33c896b9",
424
- "item_j": "exp_8e5983854cd1",
425
- "mean_difference": 89,
426
- "difference_se": null,
427
- "uncertainty_status": "unknown",
428
- "comparison_basis": "reported_value",
429
- "comparable": 1
430
- },
431
- {
432
- "item_i": "exp_066e33c896b9",
433
- "item_j": "exp_343b2ac1c9a7",
434
- "mean_difference": 56,
435
- "difference_se": null,
436
- "uncertainty_status": "unknown",
437
- "comparison_basis": "reported_value",
438
- "comparable": 1
439
- },
440
- {
441
- "item_i": "exp_066e33c896b9",
442
- "item_j": "exp_e01ea673fa5d",
443
- "mean_difference": 4,
444
- "difference_se": null,
445
- "uncertainty_status": "unknown",
446
- "comparison_basis": "reported_value",
447
- "comparable": 1
448
- },
449
- {
450
- "item_i": "exp_066e33c896b9",
451
- "item_j": "exp_ea300526f709",
452
- "mean_difference": 50,
453
- "difference_se": null,
454
- "uncertainty_status": "unknown",
455
- "comparison_basis": "reported_value",
456
- "comparable": 1
457
- },
458
- {
459
- "item_i": "exp_fba92a14e6bb",
460
- "item_j": "exp_9279aa7cd389",
461
- "mean_difference": -28,
462
- "difference_se": null,
463
- "uncertainty_status": "unknown",
464
- "comparison_basis": "reported_value",
465
- "comparable": 1
466
- },
467
- {
468
- "item_i": "exp_fba92a14e6bb",
469
- "item_j": "exp_8e5983854cd1",
470
- "mean_difference": 43,
471
- "difference_se": null,
472
- "uncertainty_status": "unknown",
473
- "comparison_basis": "reported_value",
474
- "comparable": 1
475
- },
476
- {
477
- "item_i": "exp_fba92a14e6bb",
478
- "item_j": "exp_343b2ac1c9a7",
479
- "mean_difference": 10,
480
- "difference_se": null,
481
- "uncertainty_status": "unknown",
482
- "comparison_basis": "reported_value",
483
- "comparable": 1
484
- },
485
- {
486
- "item_i": "exp_fba92a14e6bb",
487
- "item_j": "exp_e01ea673fa5d",
488
- "mean_difference": -42,
489
- "difference_se": null,
490
- "uncertainty_status": "unknown",
491
- "comparison_basis": "reported_value",
492
- "comparable": 1
493
- },
494
- {
495
- "item_i": "exp_fba92a14e6bb",
496
- "item_j": "exp_ea300526f709",
497
- "mean_difference": 4,
498
- "difference_se": null,
499
- "uncertainty_status": "unknown",
500
- "comparison_basis": "reported_value",
501
- "comparable": 1
502
- },
503
- {
504
- "item_i": "exp_9279aa7cd389",
505
- "item_j": "exp_8e5983854cd1",
506
- "mean_difference": 71,
507
- "difference_se": null,
508
- "uncertainty_status": "unknown",
509
- "comparison_basis": "reported_value",
510
- "comparable": 1
511
- },
512
- {
513
- "item_i": "exp_9279aa7cd389",
514
- "item_j": "exp_343b2ac1c9a7",
515
- "mean_difference": 38,
516
- "difference_se": null,
517
- "uncertainty_status": "unknown",
518
- "comparison_basis": "reported_value",
519
- "comparable": 1
520
- },
521
- {
522
- "item_i": "exp_9279aa7cd389",
523
- "item_j": "exp_e01ea673fa5d",
524
- "mean_difference": -14,
525
- "difference_se": null,
526
- "uncertainty_status": "unknown",
527
- "comparison_basis": "reported_value",
528
- "comparable": 1
529
- },
530
- {
531
- "item_i": "exp_9279aa7cd389",
532
- "item_j": "exp_ea300526f709",
533
- "mean_difference": 32,
534
- "difference_se": null,
535
- "uncertainty_status": "unknown",
536
- "comparison_basis": "reported_value",
537
- "comparable": 1
538
- },
539
- {
540
- "item_i": "exp_8e5983854cd1",
541
- "item_j": "exp_343b2ac1c9a7",
542
- "mean_difference": -33,
543
- "difference_se": null,
544
- "uncertainty_status": "unknown",
545
- "comparison_basis": "reported_value",
546
- "comparable": 1
547
- },
548
- {
549
- "item_i": "exp_8e5983854cd1",
550
- "item_j": "exp_e01ea673fa5d",
551
- "mean_difference": -85,
552
- "difference_se": null,
553
- "uncertainty_status": "unknown",
554
- "comparison_basis": "reported_value",
555
- "comparable": 1
556
- },
557
- {
558
- "item_i": "exp_8e5983854cd1",
559
- "item_j": "exp_ea300526f709",
560
- "mean_difference": -39,
561
- "difference_se": null,
562
- "uncertainty_status": "unknown",
563
- "comparison_basis": "reported_value",
564
- "comparable": 1
565
- },
566
- {
567
- "item_i": "exp_343b2ac1c9a7",
568
- "item_j": "exp_e01ea673fa5d",
569
- "mean_difference": -52,
570
- "difference_se": null,
571
- "uncertainty_status": "unknown",
572
- "comparison_basis": "reported_value",
573
- "comparable": 1
574
- },
575
- {
576
- "item_i": "exp_343b2ac1c9a7",
577
- "item_j": "exp_ea300526f709",
578
- "mean_difference": -6,
579
- "difference_se": null,
580
- "uncertainty_status": "unknown",
581
- "comparison_basis": "reported_value",
582
- "comparable": 1
583
- },
584
- {
585
- "item_i": "exp_e01ea673fa5d",
586
- "item_j": "exp_ea300526f709",
587
- "mean_difference": 46,
588
- "difference_se": null,
589
- "uncertainty_status": "unknown",
590
- "comparison_basis": "reported_value",
591
- "comparable": 1
592
- }
593
- ],
594
- "comparable_pairs": 21,
595
- "quality_status": "passed"
596
  },
597
  {
598
  "group_id": "g3",
599
  "items": [
600
  {
601
- "item_id": "exp_529d2754aa6a",
602
- "mean": 93,
603
  "n": null,
604
  "uncertainty_status": "unknown",
605
- "uncertainty_scale": null,
606
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
607
- "quality_status": "retained_with_flag",
608
- "warnings": [
609
- "Independent-mean uncertainty is unknown."
610
- ],
611
- "source_location": "Original-SI.pdf physical p12 FigureS2 scheme andcompleteprocedure",
612
- "aggregation": "Published endpoint value without reweighting or selection of readings."
613
  },
614
  {
615
- "item_id": "exp_94c775c85236",
616
- "mean": 90,
617
  "n": null,
618
  "uncertainty_status": "unknown",
619
- "uncertainty_scale": null,
620
- "uncertainty_basis": "No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.",
621
- "quality_status": "retained_with_flag",
622
- "warnings": [
623
- "Independent-mean uncertainty is unknown."
624
- ],
625
- "source_location": "Original-SI.pdf physical p12 FigureS2 scheme andcompleteprocedure",
626
- "aggregation": "Published endpoint value without reweighting or selection of readings."
627
- }
628
- ],
629
- "pairs": [
630
- {
631
- "item_i": "exp_529d2754aa6a",
632
- "item_j": "exp_94c775c85236",
633
- "mean_difference": 3,
634
- "difference_se": null,
635
- "uncertainty_status": "unknown",
636
- "comparison_basis": "reported_value",
637
- "comparable": 1
638
  }
639
- ],
640
- "comparable_pairs": 1,
641
- "quality_status": "passed"
642
  }
643
  ]
644
  }
 
1
  {
2
  "task_id": "alkene-transfer-conditions-ranking",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "g1",
7
  "items": [
8
  {
9
+ "item_id": "exp_6c7c207a3e35",
10
+ "mean": 95,
11
  "n": null,
12
  "uncertainty_status": "unknown",
13
+ "uncertainty_scale": null
 
 
 
 
 
 
 
14
  },
15
  {
16
+ "item_id": "exp_777d484cc14b",
17
  "mean": 89,
18
  "n": null,
19
  "uncertainty_status": "unknown",
20
+ "uncertainty_scale": null
 
 
 
 
 
 
 
21
  },
22
  {
23
+ "item_id": "exp_7ee0f9400be1",
24
  "mean": 66,
25
  "n": null,
26
  "uncertainty_status": "unknown",
27
+ "uncertainty_scale": null
 
 
 
 
 
 
 
28
  },
29
  {
30
+ "item_id": "exp_b25afc3d08f5",
31
  "mean": 11,
32
  "n": null,
33
  "uncertainty_status": "unknown",
34
+ "uncertainty_scale": null
 
 
 
 
 
 
 
35
  },
36
  {
37
+ "item_id": "exp_ecccb58cf20b",
38
+ "mean": 97,
39
  "n": null,
40
  "uncertainty_status": "unknown",
41
+ "uncertainty_scale": null
 
 
 
 
 
 
 
42
  },
43
  {
44
+ "item_id": "exp_f6a85f2013bc",
45
  "mean": 53,
46
  "n": null,
47
  "uncertainty_status": "unknown",
48
+ "uncertainty_scale": null
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
49
  }
50
+ ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
51
  },
52
  {
53
  "group_id": "g2",
54
  "items": [
55
  {
56
+ "item_id": "exp_039643486694",
57
+ "mean": 48,
58
  "n": null,
59
  "uncertainty_status": "unknown",
60
+ "uncertainty_scale": null
 
 
 
 
 
 
 
61
  },
62
  {
63
+ "item_id": "exp_064708d49167",
64
+ "mean": 90,
65
  "n": null,
66
  "uncertainty_status": "unknown",
67
+ "uncertainty_scale": null
 
 
 
 
 
 
 
68
  },
69
  {
70
+ "item_id": "exp_121a21fd986b",
71
+ "mean": 94,
72
  "n": null,
73
  "uncertainty_status": "unknown",
74
+ "uncertainty_scale": null
 
 
 
 
 
 
 
75
  },
76
  {
77
+ "item_id": "exp_3c5d2113087d",
78
  "mean": 5,
79
  "n": null,
80
  "uncertainty_status": "unknown",
81
+ "uncertainty_scale": null
 
 
 
 
 
 
 
82
  },
83
  {
84
+ "item_id": "exp_5fe231ef098e",
85
  "mean": 38,
86
  "n": null,
87
  "uncertainty_status": "unknown",
88
+ "uncertainty_scale": null
 
 
 
 
 
 
 
89
  },
90
  {
91
+ "item_id": "exp_a3d319499eac",
92
+ "mean": 76,
93
  "n": null,
94
  "uncertainty_status": "unknown",
95
+ "uncertainty_scale": null
 
 
 
 
 
 
 
96
  },
97
  {
98
+ "item_id": "exp_dddc25cdd5a3",
99
  "mean": 44,
100
  "n": null,
101
  "uncertainty_status": "unknown",
102
+ "uncertainty_scale": null
 
 
 
 
 
 
 
103
  }
104
+ ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
105
  },
106
  {
107
  "group_id": "g3",
108
  "items": [
109
  {
110
+ "item_id": "exp_b167406216a1",
111
+ "mean": 90,
112
  "n": null,
113
  "uncertainty_status": "unknown",
114
+ "uncertainty_scale": null
 
 
 
 
 
 
 
115
  },
116
  {
117
+ "item_id": "exp_b81cc665557a",
118
+ "mean": 93,
119
  "n": null,
120
  "uncertainty_status": "unknown",
121
+ "uncertainty_scale": null
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
122
  }
123
+ ]
 
 
124
  }
125
  ]
126
  }
tasks/alkene-transfer-conditions-ranking/tests/contract.json CHANGED
@@ -1,72 +1,59 @@
1
  {
2
  "task_id": "alkene-transfer-conditions-ranking",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "g1",
7
  "label": "4-Phenyl-1-butene cross-metathesis adduct yield",
8
  "item_ids": [
9
- "exp_41bd98dba742",
10
- "exp_484d7efaebd5",
11
- "exp_389c830f7e89",
12
- "exp_dd127eb2b1f5",
13
- "exp_8545262bd121",
14
- "exp_369ffbca2e28",
15
- "exp_cac98fcd26c1"
16
  ],
17
  "direction": "ascending",
18
  "group_weight": 0.3333333333333333,
19
- "k": 2,
20
- "epsilon": 0,
21
- "uncertainty_model": "optional_reported_uncertainty",
22
- "difference_standard_error": "sem_sum_upper_bound",
23
  "endpoint": {
24
  "name": "product 1H NMR yield",
25
  "unit": "%",
26
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
27
  }
28
  },
29
  {
30
  "group_id": "g2",
31
  "label": "Rotenone cross-metathesis adduct yield",
32
  "item_ids": [
33
- "exp_066e33c896b9",
34
- "exp_fba92a14e6bb",
35
- "exp_9279aa7cd389",
36
- "exp_8e5983854cd1",
37
- "exp_343b2ac1c9a7",
38
- "exp_e01ea673fa5d",
39
- "exp_ea300526f709"
40
  ],
41
  "direction": "ascending",
42
  "group_weight": 0.3333333333333333,
43
- "k": 2,
44
- "epsilon": 0,
45
- "uncertainty_model": "optional_reported_uncertainty",
46
- "difference_standard_error": "sem_sum_upper_bound",
47
  "endpoint": {
48
  "name": "product 1H NMR yield",
49
  "unit": "%",
50
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
51
  }
52
  },
53
  {
54
  "group_id": "g3",
55
  "label": "Combined alkene yield from E/Z latent precursors",
56
  "item_ids": [
57
- "exp_529d2754aa6a",
58
- "exp_94c775c85236"
59
  ],
60
  "direction": "ascending",
61
  "group_weight": 0.3333333333333333,
62
- "k": 2,
63
- "epsilon": 0,
64
- "uncertainty_model": "optional_reported_uncertainty",
65
- "difference_standard_error": "sem_sum_upper_bound",
66
  "endpoint": {
67
  "name": "product 1H NMR yield",
68
  "unit": "%",
69
- "definition_en": "Reported1H NMR yield of the group-specified product relative to initial alkene or latent precursor,using methyl benzoate internal standard."
70
  }
71
  }
72
  ],
@@ -79,18 +66,9 @@
79
  "spearman_tie_method": "average",
80
  "spearman_undefined_policy": "task_error",
81
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
82
- "secondary_scoring_method": "equal_pair_inversion_random95",
83
- "ranking_reward_range": [
84
- 0,
85
- 1
86
- ],
87
- "random_quantile": 0.95,
88
- "unknown_uncertainty_policy": "use_reported_value",
89
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
90
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
91
  "format_errors": {
92
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
93
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
94
  },
95
  "max_submission_bytes": 1048576,
96
  "minimum_conditions_per_group": 2,
@@ -107,5 +85,14 @@
107
  "report_path": "/logs/artifacts/report.md",
108
  "sources_path": "/logs/artifacts/sources.json",
109
  "analysis_path": "/logs/artifacts/analysis/",
110
- "forbidden_result_scope": "Measured outcomes, yields, rankings, equivalent endpoints and result-fitted predictors for alkene transfer-reagent and catalyst conditions in the stated material and reaction system are prohibited."
 
 
 
 
 
 
 
 
 
111
  }
 
1
  {
2
  "task_id": "alkene-transfer-conditions-ranking",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "g1",
7
  "label": "4-Phenyl-1-butene cross-metathesis adduct yield",
8
  "item_ids": [
9
+ "exp_6c7c207a3e35",
10
+ "exp_777d484cc14b",
11
+ "exp_7ee0f9400be1",
12
+ "exp_b25afc3d08f5",
13
+ "exp_ecccb58cf20b",
14
+ "exp_f6a85f2013bc"
 
15
  ],
16
  "direction": "ascending",
17
  "group_weight": 0.3333333333333333,
 
 
 
 
18
  "endpoint": {
19
  "name": "product 1H NMR yield",
20
  "unit": "%",
21
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
22
  }
23
  },
24
  {
25
  "group_id": "g2",
26
  "label": "Rotenone cross-metathesis adduct yield",
27
  "item_ids": [
28
+ "exp_039643486694",
29
+ "exp_064708d49167",
30
+ "exp_121a21fd986b",
31
+ "exp_3c5d2113087d",
32
+ "exp_5fe231ef098e",
33
+ "exp_a3d319499eac",
34
+ "exp_dddc25cdd5a3"
35
  ],
36
  "direction": "ascending",
37
  "group_weight": 0.3333333333333333,
 
 
 
 
38
  "endpoint": {
39
  "name": "product 1H NMR yield",
40
  "unit": "%",
41
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
42
  }
43
  },
44
  {
45
  "group_id": "g3",
46
  "label": "Combined alkene yield from E/Z latent precursors",
47
  "item_ids": [
48
+ "exp_b167406216a1",
49
+ "exp_b81cc665557a"
50
  ],
51
  "direction": "ascending",
52
  "group_weight": 0.3333333333333333,
 
 
 
 
53
  "endpoint": {
54
  "name": "product 1H NMR yield",
55
  "unit": "%",
56
+ "definition_en": "Reported 1H NMR yield of the group-specified product relative to initial alkene or latent precursor, using methyl benzoate internal standard."
57
  }
58
  }
59
  ],
 
66
  "spearman_tie_method": "average",
67
  "spearman_undefined_policy": "task_error",
68
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
69
  "format_errors": {
70
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
71
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
72
  },
73
  "max_submission_bytes": 1048576,
74
  "minimum_conditions_per_group": 2,
 
85
  "report_path": "/logs/artifacts/report.md",
86
  "sources_path": "/logs/artifacts/sources.json",
87
  "analysis_path": "/logs/artifacts/analysis/",
88
+ "forbidden_result_scope": "Measured outcomes, yields, rankings, equivalent endpoints and result-fitted predictors for alkene transfer-reagent and catalyst conditions in the stated material and reaction system are prohibited.",
89
+ "supporting_artifact_review": {
90
+ "policy_stop_marker": "SPU_POLICY_STOP",
91
+ "required_files": [
92
+ "report.md",
93
+ "sources.json"
94
+ ],
95
+ "required_directory": "analysis",
96
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
97
+ }
98
  }
tasks/alkene-transfer-conditions-ranking/tests/grade.py CHANGED
@@ -1,14 +1,44 @@
1
- """Pipeline v18 Spearman core and preserved equal-pair auxiliary grader."""
2
  import argparse
3
- import copy
4
  import hashlib
5
  import json
6
  import math
 
 
7
  from pathlib import Path
8
 
9
- import original_grade as old
10
 
11
- TaskError = old.TaskError
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
 
13
 
14
  def bounded(value):
@@ -51,98 +81,174 @@ def spearman(values, predicted_positions):
51
  return bounded(math.fsum(r * p for r, p in zip(x, y)) / denominator)
52
 
53
 
54
- def promote_contract(contract):
55
- value = copy.deepcopy(contract)
56
- value.update(scoring_method='spearman', primary_score='task_spearman',
57
- spearman_range=[-1, 1], spearman_tie_method='average',
58
- spearman_undefined_policy='task_error',
59
- secondary_scoring_method='equal_pair_inversion_random95',
60
- ranking_reward_range=[0, 1])
61
- return value
62
-
63
-
64
- def auxiliary_view(contract):
65
- expected = {'scoring_method': 'spearman', 'primary_score': 'task_spearman',
66
- 'spearman_range': [-1, 1], 'spearman_tie_method': 'average',
67
- 'spearman_undefined_policy': 'task_error',
68
- 'secondary_scoring_method': 'equal_pair_inversion_random95',
69
- 'ranking_reward_range': [0, 1]}
70
- if any(contract.get(k) != v for k, v in expected.items()):
71
- raise TaskError('Invalid dual-score contract')
72
- value = copy.deepcopy(contract)
73
- value['scoring_method'] = value.pop('secondary_scoring_method')
74
- for key in ('primary_score', 'spearman_range', 'spearman_tie_method',
75
- 'spearman_undefined_policy', 'ranking_reward_range'):
76
- value.pop(key)
77
- return value
78
-
79
-
80
- def scoring_bindings(contract, answers):
81
- return {'contract_sha256': old.canonical_sha(contract),
82
- 'answers_sha256': old.canonical_sha(answers),
83
- 'grader_sha256': hashlib.sha256(Path(__file__).read_bytes()).hexdigest(),
84
- 'auxiliary_grader_sha256': hashlib.sha256(Path(old.__file__).read_bytes()).hexdigest()}
85
-
86
-
87
- def validate(contract, answers, baseline):
88
- auxiliary = auxiliary_view(contract)
89
- old.validate_task(auxiliary, answers)
90
- if baseline.get('bindings') != scoring_bindings(contract, answers):
91
- raise TaskError('Dual-score baseline hash mismatch')
92
- old_baseline = {'q': baseline.get('q'), 'bindings': old.scoring_bindings(auxiliary, answers)}
93
- old.validate_baseline(auxiliary, answers, old_baseline)
94
- hidden = {g['group_id']: {i['item_id']: i['mean'] for i in g['items']} for g in answers['groups']}
95
- for g in contract['groups']:
96
- spearman([hidden[g['group_id']][i] for i in g['item_ids']], list(range(len(g['item_ids']))))
97
- return auxiliary, old_baseline, hidden
98
-
99
-
100
- def add_core(contract, report, hidden):
101
- report['scoring_method'] = 'spearman'
102
- report['primary_score'] = 'task_spearman'
103
- report['secondary_scoring_method'] = 'equal_pair_inversion_random95'
104
- report['official_reward'] = None
105
- report['official_ranking_reward'] = None
106
- if not report['groups']:
107
- report['task_spearman'] = -1.0
108
- report['core_penalty_reason'] = 'whole_task_format_error'
109
- report['groups'] = [{'group_id': g['group_id'], 'group_weight': g['group_weight'],
110
- 'group_spearman_g': None, 'spearman_for_aggregation': -1.0,
111
- 'group_pair_accuracy': 0.0, 'status': 'whole_task_format_error'}
112
- for g in contract['groups']]
113
- return report
114
- for g in report['groups']:
115
- if g['status'] != 'valid':
116
- rho = None
117
- else:
118
- item_ids = next(c['item_ids'] for c in contract['groups'] if c['group_id'] == g['group_id'])
119
- positions = {item_id: i for i, item_id in enumerate(g['order'])}
120
- rho = spearman([hidden[g['group_id']][i] for i in item_ids], [positions[i] for i in item_ids])
121
- g['group_spearman_g'] = rho
122
- g['spearman_for_aggregation'] = -1.0 if rho is None else rho
123
- report['task_spearman'] = bounded(math.fsum(g['group_weight'] * g['spearman_for_aggregation']
124
- for g in report['groups']))
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
125
  return report
126
 
127
 
128
- def grade_submission(contract, answers, submission, baseline):
129
- auxiliary, old_baseline, hidden = validate(contract, answers, baseline)
130
- report = old.grade_submission(auxiliary, answers, submission, old_baseline)
131
- report['random_q95'] = baseline['q']
132
- return add_core(contract, report, hidden)
133
 
134
 
135
- def grade_file(contract, answers, path, baseline):
136
- auxiliary, old_baseline, hidden = validate(contract, answers, baseline)
137
- return add_core(contract, old.grade_file(auxiliary, answers, path, old_baseline), hidden)
138
 
 
 
 
 
 
 
 
 
 
 
 
 
139
 
140
- def official_scores(report, compliance, execution, memory=False):
141
- if compliance == 'cheating':
142
- return -1.0, 0.0
143
- if compliance != 'compliant' or execution != 'completed':
144
- return None, None
145
- return report['task_spearman'], report['ranking_reward']
 
 
 
 
 
 
 
146
 
147
 
148
  def main():
@@ -156,18 +262,22 @@ def main():
156
  for name in ('reward.txt', 'reward.json', 'report.json', 'evaluator_error.json'):
157
  (args.output / name).unlink(missing_ok=True)
158
  try:
159
- contract, answers, baseline = [old.load_json((args.root / f'{name}.json').read_bytes())
160
- for name in ('contract', 'answers', 'random_baseline')]
161
- report = grade_file(contract, answers, args.submission, baseline)
162
- except (TaskError, old.SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
163
- old.dump(args.output / 'evaluator_error.json', {'status': 'task_error', 'error': str(error)})
 
 
 
 
164
  raise SystemExit(2) from error
165
- old.dump(args.output / 'report.json', report)
 
166
  if args.interface == 'txt':
167
  (args.output / 'reward.txt').write_text(str(report['task_spearman']) + '\n')
168
  else:
169
- old.dump(args.output / 'reward.json', {'reward': report['task_spearman'],
170
- 'ranking_reward': report['ranking_reward']})
171
 
172
 
173
  if __name__ == '__main__':
 
1
+ """Score complete experimental rankings by weighted Spearman correlation."""
2
  import argparse
3
+ import errno
4
  import hashlib
5
  import json
6
  import math
7
+ import os
8
+ import stat
9
  from pathlib import Path
10
 
 
11
 
12
+ class TaskError(ValueError):
13
+ pass
14
+
15
+
16
+ class SubmissionError(ValueError):
17
+ pass
18
+
19
+
20
+ def unique_object(pairs):
21
+ result = {}
22
+ for key, value in pairs:
23
+ if key in result:
24
+ raise SubmissionError("duplicate_json_key")
25
+ result[key] = value
26
+ return result
27
+
28
+
29
+ def load_json(raw):
30
+ def reject_constant(value):
31
+ raise SubmissionError("nonfinite_json_constant")
32
+ return json.loads(raw.decode("utf-8"), object_pairs_hook=unique_object,
33
+ parse_constant=reject_constant)
34
+
35
+
36
+ def finite(value, name, minimum=None):
37
+ if isinstance(value, bool) or not isinstance(value, (int, float)):
38
+ raise TaskError(name + " is not numeric")
39
+ if not math.isfinite(value) or (minimum is not None and value < minimum):
40
+ raise TaskError(name + " is out of range")
41
+ return float(value)
42
 
43
 
44
  def bounded(value):
 
81
  return bounded(math.fsum(r * p for r, p in zip(x, y)) / denominator)
82
 
83
 
84
+ def read_regular(path, limit):
85
+ try:
86
+ fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
87
+ try:
88
+ info = os.fstat(fd)
89
+ if not stat.S_ISREG(info.st_mode):
90
+ raise SubmissionError("not_regular_file")
91
+ if info.st_size > limit:
92
+ raise SubmissionError("submission_too_large")
93
+ with os.fdopen(fd, "rb", closefd=False) as stream:
94
+ raw = stream.read(limit + 1)
95
+ if len(raw) > limit:
96
+ raise SubmissionError("submission_too_large")
97
+ return raw
98
+ finally:
99
+ os.close(fd)
100
+ except OSError as error:
101
+ if error.errno in (errno.ENOENT, errno.ELOOP, errno.EISDIR, errno.ENOTDIR,
102
+ errno.ENXIO, errno.EACCES, errno.EPERM):
103
+ raise SubmissionError("missing_or_unreadable_submission") from error
104
+ raise
105
+
106
+
107
+ def validate_task(contract, answers):
108
+ expected = {
109
+ "scoring_method": "spearman",
110
+ "primary_score": "task_spearman",
111
+ "spearman_range": [-1, 1],
112
+ "spearman_tie_method": "average",
113
+ "spearman_undefined_policy": "task_error",
114
+ }
115
+ if any(contract.get(key) != value for key, value in expected.items()):
116
+ raise TaskError("Unsupported scoring contract")
117
+ for key in ("task_id", "task_version"):
118
+ if not isinstance(contract.get(key), str) or answers.get(key) != contract[key]:
119
+ raise TaskError("Invalid task identity")
120
+ limit = contract.get("max_submission_bytes")
121
+ if type(limit) is not int or not 0 < limit <= 1048576:
122
+ raise TaskError("Invalid submission size limit")
123
+ minimum = contract.get("minimum_conditions_per_group", 2)
124
+ if type(minimum) is not int or minimum < 2:
125
+ raise TaskError("Invalid minimum group size")
126
+ groups = contract["groups"]
127
+ references = {g["group_id"]: g for g in answers["groups"]}
128
+ if not groups or len(groups) != len(references) or len(references) != len(answers["groups"]):
129
+ raise TaskError("Invalid reference groups")
130
+
131
+ measurements = {}
132
+ all_ids = set()
133
+ for group in groups:
134
+ gid = group["group_id"]
135
+ ids = group["item_ids"]
136
+ if not isinstance(gid, str) or not gid or gid in measurements or gid not in references:
137
+ raise TaskError("Invalid group ID")
138
+ if (group["direction"] != "ascending" or not isinstance(ids, list) or len(ids) < minimum
139
+ or any(not isinstance(i, str) or not i for i in ids)
140
+ or len(set(ids)) != len(ids) or all_ids.intersection(ids)):
141
+ raise TaskError("Invalid group definition")
142
+ all_ids.update(ids)
143
+ if finite(group["group_weight"], "group_weight", 0) == 0:
144
+ raise TaskError("Zero group weight")
145
+ items = references[gid]["items"]
146
+ values = {item["item_id"]: item["mean"] for item in items}
147
+ if len(items) != len(ids) or set(values) != set(ids):
148
+ raise TaskError("Reference items differ from the contract")
149
+ measurements[gid] = [values[i] for i in ids]
150
+ spearman(measurements[gid], list(range(len(ids))))
151
+ if not math.isclose(math.fsum(g["group_weight"] for g in groups), 1.0, rel_tol=0, abs_tol=1e-12):
152
+ raise TaskError("Group weights must sum to one")
153
+ return measurements
154
+
155
+
156
+ def group_entries(submission, contract):
157
+ if not isinstance(submission, dict):
158
+ raise SubmissionError("invalid_top_level_structure")
159
+ if any(submission.get(key) != contract[key] for key in ("task_id", "task_version")):
160
+ raise SubmissionError("task_identity_mismatch")
161
+ if not isinstance(submission.get("rankings"), list):
162
+ raise SubmissionError("rankings_not_array")
163
+ entries = {group["group_id"]: [] for group in contract["groups"]}
164
+ for entry in submission["rankings"]:
165
+ if not isinstance(entry, dict) or not isinstance(entry.get("group_id"), str):
166
+ raise SubmissionError("unassignable_group_entry")
167
+ if entry["group_id"] not in entries:
168
+ raise SubmissionError("unknown_group_id")
169
+ entries[entry["group_id"]].append(entry)
170
+ return entries
171
+
172
+
173
+ def score_group(group, entries, values):
174
+ gid = group["group_id"]
175
+ ids = group["item_ids"]
176
+ result = {"group_id": gid, "group_weight": group["group_weight"],
177
+ "group_spearman_g": None, "spearman_for_aggregation": -1.0}
178
+ if entries is None:
179
+ result["status"] = "whole_task_format_error"
180
+ return result
181
+ matches = entries[gid]
182
+ if len(matches) != 1:
183
+ result["status"] = "duplicate_group" if matches else "missing_group"
184
+ return result
185
+ order = matches[0].get("order")
186
+ if (not isinstance(order, list) or len(order) != len(ids)
187
+ or any(not isinstance(i, str) for i in order) or set(order) != set(ids)):
188
+ result["status"] = "invalid_permutation"
189
+ return result
190
+ positions = {item: p for p, item in enumerate(order)}
191
+ rho = spearman(values, [positions[i] for i in ids])
192
+ result.update(status="valid", order=order, group_spearman_g=rho, spearman_for_aggregation=rho)
193
+ return result
194
+
195
+
196
+ def grade_file(contract, answers, path):
197
+ measurements = validate_task(contract, answers)
198
+ report = {"task_id": contract["task_id"], "task_version": contract["task_version"],
199
+ "scoring_method": "spearman", "primary_score": "task_spearman",
200
+ "evaluation_status": "completed"}
201
+ entries = None
202
+ try:
203
+ raw = read_regular(path, contract["max_submission_bytes"])
204
+ submission = load_json(raw)
205
+ report["submission_sha256"] = hashlib.sha256(raw).hexdigest()
206
+ entries = group_entries(submission, contract)
207
+ except (SubmissionError, UnicodeError, json.JSONDecodeError, RecursionError) as error:
208
+ report["submission_status"] = str(error) or type(error).__name__
209
+ report["core_penalty_reason"] = "whole_task_format_error"
210
+ groups = [score_group(g, entries, measurements[g["group_id"]]) for g in contract["groups"]]
211
+ report["groups"] = groups
212
+ if entries is None:
213
+ report["task_spearman"] = -1.0
214
+ else:
215
+ report["submission_status"] = "valid" if all(g["status"] == "valid" for g in groups) else "group_errors"
216
+ report["task_spearman"] = bounded(math.fsum(g["group_weight"] * g["spearman_for_aggregation"]
217
+ for g in groups))
218
  return report
219
 
220
 
221
+ def dump(path, value):
222
+ path.write_text(json.dumps(value, indent=2, allow_nan=False) + "\n", encoding="utf-8")
 
 
 
223
 
224
 
 
 
 
225
 
226
+ def supporting_artifact_status(submission):
227
+ root = submission.parent
228
+ report_path = root / 'report.md'
229
+ paths = {'report_file': report_path, 'sources_file': root / 'sources.json',
230
+ 'analysis_directory': root / 'analysis'}
231
+ present = {}
232
+ for name, path in paths.items():
233
+ try:
234
+ mode = path.lstat().st_mode
235
+ present[name] = stat.S_ISDIR(mode) if name == 'analysis_directory' else stat.S_ISREG(mode)
236
+ except OSError:
237
+ present[name] = False
238
 
239
+ declared_stop = False
240
+ if present['report_file']:
241
+ try:
242
+ fd = os.open(report_path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
243
+ with os.fdopen(fd, 'rb') as stream:
244
+ if stat.S_ISREG(os.fstat(stream.fileno()).st_mode):
245
+ declared_stop = stream.readline(256).rstrip(b'\r\n') == b'SPU_POLICY_STOP'
246
+ except OSError:
247
+ pass
248
+ status = 'artifacts_present' if all(present.values()) else 'artifacts_incomplete'
249
+ if declared_stop:
250
+ status = 'declared_stop_with_ranking' if os.path.lexists(submission) else 'declared_stop'
251
+ return {'status': status, **present, 'policy_stop_declared': declared_stop}
252
 
253
 
254
  def main():
 
262
  for name in ('reward.txt', 'reward.json', 'report.json', 'evaluator_error.json'):
263
  (args.output / name).unlink(missing_ok=True)
264
  try:
265
+ bindings = load_json((args.root / 'scoring_hashes.json').read_bytes())
266
+ for name in ('contract.json', 'answers.json', 'grade.py'):
267
+ if hashlib.sha256((args.root / name).read_bytes()).hexdigest() != bindings[name]:
268
+ raise TaskError('Scoring file hash mismatch: ' + name)
269
+ contract = load_json((args.root / 'contract.json').read_bytes())
270
+ answers = load_json((args.root / 'answers.json').read_bytes())
271
+ report = grade_file(contract, answers, args.submission)
272
+ except (TaskError, SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
273
+ dump(args.output / 'evaluator_error.json', {'status': 'task_error', 'error': str(error)})
274
  raise SystemExit(2) from error
275
+ report['supporting_artifact_review'] = supporting_artifact_status(args.submission)
276
+ dump(args.output / 'report.json', report)
277
  if args.interface == 'txt':
278
  (args.output / 'reward.txt').write_text(str(report['task_spearman']) + '\n')
279
  else:
280
+ dump(args.output / 'reward.json', {'reward': report['task_spearman']})
 
281
 
282
 
283
  if __name__ == '__main__':
tasks/alkene-transfer-conditions-ranking/tests/measurements.csv CHANGED
@@ -1,17 +1,16 @@
1
  group_id,item_id,mean,n,uncertainty_status,uncertainty_scale,uncertainty_basis,quality_status,warnings,source_location,aggregation
2
- g1,exp_41bd98dba742,97,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS3 entry2,catalyst structures;p5section3.1method",Published endpoint value without reweighting or selection of readings.
3
- g1,exp_484d7efaebd5,89,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p5 TableS1 entry2,scheme andsection3.1method",Published endpoint value without reweighting or selection of readings.
4
- g1,exp_389c830f7e89,66,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p5 TableS1 entry1,scheme andsection3.1method",Published endpoint value without reweighting or selection of readings.
5
- g1,exp_dd127eb2b1f5,11,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS3 entry3,catalyst structures;p5section3.1method",Published endpoint value without reweighting or selection of readings.
6
- g1,exp_8545262bd121,53,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS4 entry3,scheme;p5section3.1method",Published endpoint value without reweighting or selection of readings.
7
- g1,exp_369ffbca2e28,53,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS4 entry2,scheme;p5section3.1method",Published endpoint value without reweighting or selection of readings.
8
- g1,exp_cac98fcd26c1,95,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p5 TableS1 entry3,scheme andsection3.1method",Published endpoint value without reweighting or selection of readings.
9
- g2,exp_066e33c896b9,94,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry7,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
10
- g2,exp_fba92a14e6bb,48,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry2,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
11
- g2,exp_9279aa7cd389,76,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry6,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
12
- g2,exp_8e5983854cd1,5,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry5,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
13
- g2,exp_343b2ac1c9a7,38,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry3,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
14
- g2,exp_e01ea673fa5d,90,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry8,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
15
- g2,exp_ea300526f709,44,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry1,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
16
- g3,exp_529d2754aa6a,93,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]",Original-SI.pdf physical p12 FigureS2 scheme andcompleteprocedure,Published endpoint value without reweighting or selection of readings.
17
- g3,exp_94c775c85236,90,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]",Original-SI.pdf physical p12 FigureS2 scheme andcompleteprocedure,Published endpoint value without reweighting or selection of readings.
 
1
  group_id,item_id,mean,n,uncertainty_status,uncertainty_scale,uncertainty_basis,quality_status,warnings,source_location,aggregation
2
+ g1,exp_6c7c207a3e35,95,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p5 TableS1 entry3,scheme andsection3.1method",Published endpoint value without reweighting or selection of readings.
3
+ g1,exp_777d484cc14b,89,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p5 TableS1 entry2,scheme andsection3.1method",Published endpoint value without reweighting or selection of readings.
4
+ g1,exp_7ee0f9400be1,66,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p5 TableS1 entry1,scheme andsection3.1method",Published endpoint value without reweighting or selection of readings.
5
+ g1,exp_b25afc3d08f5,11,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS3 entry3,catalyst structures;p5section3.1method",Published endpoint value without reweighting or selection of readings.
6
+ g1,exp_ecccb58cf20b,97,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS3 entry2,catalyst structures;p5section3.1method",Published endpoint value without reweighting or selection of readings.
7
+ g1,exp_f6a85f2013bc,53,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p6 TableS4 entry3,scheme;p5section3.1method",Published endpoint value without reweighting or selection of readings.
8
+ g2,exp_039643486694,48,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry2,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
9
+ g2,exp_064708d49167,90,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry8,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
10
+ g2,exp_121a21fd986b,94,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry7,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
11
+ g2,exp_3c5d2113087d,5,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry5,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
12
+ g2,exp_5fe231ef098e,38,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry3,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
13
+ g2,exp_a3d319499eac,76,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry6,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
14
+ g2,exp_dddc25cdd5a3,44,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]","Original-SI.pdf physical p9 TableS6 entry1,reaction/catalyststructures andsection3.3procedure",Published endpoint value without reweighting or selection of readings.
15
+ g3,exp_b167406216a1,90,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]",Original-SI.pdf physical p12 FigureS2 scheme andcompleteprocedure,Published endpoint value without reweighting or selection of readings.
16
+ g3,exp_b81cc665557a,93,,unknown,,No reliable independent-mean uncertainty is supplied. Unknown is not treated as zero.,retained_with_flag,"[""Independent-mean uncertainty is unknown.""]",Original-SI.pdf physical p12 FigureS2 scheme andcompleteprocedure,Published endpoint value without reweighting or selection of readings.
 
tasks/alkene-transfer-conditions-ranking/tests/original_grade.py DELETED
@@ -1,423 +0,0 @@
1
- """Pipeline v16: equal pairs within groups; one task-level random-95 threshold."""
2
-
3
- import argparse
4
- import errno
5
- import hashlib
6
- import itertools
7
- import json
8
- import math
9
- import os
10
- from pathlib import Path
11
- import stat
12
-
13
-
14
- class TaskError(ValueError):
15
- pass
16
-
17
-
18
- class SubmissionError(ValueError):
19
- pass
20
-
21
-
22
- def canonical_sha(value):
23
- return hashlib.sha256(json.dumps(value, sort_keys=True, separators=(",", ":"),
24
- ensure_ascii=False, allow_nan=False).encode()).hexdigest()
25
-
26
-
27
- def scoring_bindings(contract, answers):
28
- return {
29
- "contract_sha256": canonical_sha(contract),
30
- "answers_sha256": canonical_sha(answers),
31
- "relations_sha256": canonical_sha([{"group_id": g["group_id"], "pairs": g["pairs"]}
32
- for g in answers["groups"]]),
33
- "weights_sha256": canonical_sha([{k: g[k] for k in ("group_id", "group_weight")}
34
- for g in contract["groups"]]),
35
- "grader_sha256": hashlib.sha256(Path(__file__).read_bytes()).hexdigest(),
36
- }
37
-
38
-
39
- def validate_baseline(contract, answers, baseline):
40
- if not isinstance(baseline, dict):
41
- raise TaskError("random baseline must be an object")
42
- q = finite(baseline.get("q"), "random_q95", 0.5)
43
- if q >= 1:
44
- raise TaskError("random_q95 must be below 1")
45
- if contract.get("random_quantile") != 0.95:
46
- raise TaskError("invalid random quantile")
47
- if baseline.get("bindings") != scoring_bindings(contract, answers):
48
- raise TaskError("random baseline hash mismatch")
49
- return q
50
-
51
-
52
- def reward_from_accuracy(accuracy, q):
53
- accuracy = finite(accuracy, "accuracy", 0)
54
- q = finite(q, "random_q95", 0.5)
55
- if accuracy > 1 or q >= 1:
56
- raise TaskError("invalid accuracy or random threshold")
57
- if accuracy == 1:
58
- return 1.0
59
- if accuracy <= q + 1e-12:
60
- return 0.0
61
- return min(1.0, max(0.0, (accuracy - q) / (1 - q)))
62
-
63
-
64
- def aggregate_accuracy(details):
65
- # Preserve exact endpoints despite representational roundoff in weights.
66
- if all(g["group_pair_accuracy"] == 1 for g in details):
67
- return 1.0
68
- return math.fsum(g["group_pair_accuracy"] * g["group_weight"] for g in details)
69
-
70
-
71
- def unique_object(pairs):
72
- result = {}
73
- for key, value in pairs:
74
- if key in result:
75
- raise SubmissionError("duplicate_json_key")
76
- result[key] = value
77
- return result
78
-
79
-
80
- def load_json(raw):
81
- def reject_constant(value):
82
- raise SubmissionError("nonfinite_json_constant")
83
- return json.loads(raw.decode("utf-8"), object_pairs_hook=unique_object,
84
- parse_constant=reject_constant)
85
-
86
-
87
- def finite(value, name, minimum=None):
88
- if isinstance(value, bool) or not isinstance(value, (int, float)):
89
- raise TaskError(name + " is not numeric")
90
- if not math.isfinite(value) or (minimum is not None and value < minimum):
91
- raise TaskError(name + " is out of range")
92
- return float(value)
93
-
94
-
95
- def difference_se(sem_i, sem_j, method):
96
- if method == "sem_sum_upper_bound":
97
- return sem_i + sem_j
98
- if method == "independent_sem_quadrature":
99
- return math.hypot(sem_i, sem_j)
100
- raise TaskError("unsupported difference standard error method")
101
-
102
-
103
- def validate_task(contract, answers):
104
- if not isinstance(contract, dict) or not isinstance(answers, dict):
105
- raise TaskError("contract and answers must be objects")
106
- if contract.get("scoring_method") != "equal_pair_inversion_random95":
107
- raise TaskError("unsupported scoring method")
108
- if contract.get("unknown_uncertainty_policy") != "use_reported_value":
109
- raise TaskError("unsupported unknown-uncertainty policy")
110
- for key in ("task_id", "task_version"):
111
- if not isinstance(contract.get(key), str) or answers.get(key) != contract[key]:
112
- raise TaskError("invalid task identity")
113
- limit = contract.get("max_submission_bytes")
114
- if type(limit) is not int or not 0 < limit <= 1048576:
115
- raise TaskError("invalid submission size limit")
116
- minimum_conditions = contract.get("minimum_conditions_per_group", 2)
117
- if type(minimum_conditions) is not int or minimum_conditions < 2:
118
- raise TaskError("invalid minimum conditions")
119
- groups = contract.get("groups")
120
- hidden_groups = answers.get("groups")
121
- if not isinstance(groups, list) or not groups or not isinstance(hidden_groups, list):
122
- raise TaskError("missing groups")
123
- if len(groups) != len(hidden_groups):
124
- raise TaskError("hidden group count mismatch")
125
- hidden_map = {}
126
- for hidden in hidden_groups:
127
- if not isinstance(hidden, dict) or not isinstance(hidden.get("group_id"), str):
128
- raise TaskError("invalid hidden group")
129
- if hidden["group_id"] in hidden_map:
130
- raise TaskError("duplicate hidden group")
131
- hidden_map[hidden["group_id"]] = hidden
132
- validated, group_ids, all_ids = [], set(), set()
133
- for group in groups:
134
- if not isinstance(group, dict):
135
- raise TaskError("invalid contract group")
136
- gid = group.get("group_id")
137
- if not isinstance(gid, str) or not gid or gid in group_ids:
138
- raise TaskError("invalid group id")
139
- group_ids.add(gid)
140
- ids = group.get("item_ids")
141
- if (not isinstance(ids, list) or len(ids) < minimum_conditions
142
- or any(not isinstance(item, str) or not item for item in ids)
143
- or len(ids) != len(set(ids)) or all_ids.intersection(ids)):
144
- raise TaskError("invalid item ids")
145
- all_ids.update(ids)
146
- if group.get("direction") != "ascending":
147
- raise TaskError("only ascending order is supported")
148
- group_weight = finite(group.get("group_weight"), "group_weight", 0)
149
- if group_weight == 0:
150
- raise TaskError("zero group weight")
151
- k = finite(group.get("k"), "k", 0)
152
- epsilon = finite(group.get("epsilon"), "epsilon", 0)
153
- method = group.get("difference_standard_error")
154
- if method not in ("sem_sum_upper_bound", "independent_sem_quadrature", "unknown"):
155
- raise TaskError("unsupported difference standard error method")
156
- if group.get("uncertainty_model") not in ("optional_reported_uncertainty",
157
- "reported_sd_with_source_bounds", "independent_replicate_mean", "replicate_sd_upper_envelope", "unknown_replicate_hierarchy"):
158
- raise TaskError("unsupported marginal uncertainty model")
159
- hidden = hidden_map.get(gid)
160
- if hidden is None or hidden.get("quality_status") != "passed":
161
- raise TaskError("missing or invalid hidden group")
162
- items = hidden.get("items")
163
- if not isinstance(items, list) or len(items) != len(ids):
164
- raise TaskError("hidden item count mismatch")
165
- item_map = {}
166
- for item in items:
167
- if not isinstance(item, dict) or item.get("item_id") not in ids:
168
- raise TaskError("invalid hidden item")
169
- item_id = item["item_id"]
170
- if item_id in item_map or item.get("quality_status") not in ("retained", "retained_with_flag"):
171
- raise TaskError("duplicate or invalid hidden item")
172
- mean = finite(item.get("mean"), "mean")
173
- n = item.get("n")
174
- if n is not None and (type(n) is not int or n < 1):
175
- raise TaskError("invalid experimental replicate count")
176
- uncertainty_status = item.get("uncertainty_status")
177
- if uncertainty_status not in ("known", "unknown"):
178
- raise TaskError("invalid uncertainty status")
179
- uncertainty_model = group.get("uncertainty_model", "independent_replicate_mean")
180
- if uncertainty_model == "reported_sd_with_source_bounds":
181
- lo = finite(item.get("mean_lower"), "mean_lower")
182
- hi = finite(item.get("mean_upper"), "mean_upper")
183
- if lo > hi or not lo - 1e-12 <= mean <= hi + 1e-12:
184
- raise TaskError("inconsistent reported mean interval")
185
- if uncertainty_status == "unknown":
186
- if item.get("uncertainty_scale") is not None:
187
- raise TaskError("unknown uncertainty must be null")
188
- sem = None
189
- elif uncertainty_model == "optional_reported_uncertainty":
190
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
191
- if not item.get("uncertainty_basis"):
192
- raise TaskError("known uncertainty needs a stated source basis")
193
- elif uncertainty_model == "reported_sd_with_source_bounds":
194
- if n is None or n < 2:
195
- raise TaskError("reported SD needs repeat evidence")
196
- lower_sd = finite(item.get("sd_lower"), "sd_lower", 0)
197
- upper_sd = finite(item.get("sd_upper"), "sd_upper", 0)
198
- if lower_sd > upper_sd:
199
- raise TaskError("inconsistent reported SD interval")
200
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
201
- expected_scale = upper_sd + (hi - lo) / 2
202
- if not math.isclose(sem, expected_scale, rel_tol=1e-12, abs_tol=1e-12):
203
- raise TaskError("inconsistent source-bound uncertainty")
204
- elif uncertainty_model in ("independent_replicate_mean", "replicate_sd_upper_envelope"):
205
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
206
- sd = finite(item.get("sample_sd"), "sample_sd", 0)
207
- if n is None or n < 2:
208
- raise TaskError("known SD needs repeat evidence")
209
- expected_sem = sd / math.sqrt(n) if uncertainty_model == "independent_replicate_mean" else sd
210
- if not math.isclose(sem, expected_sem, rel_tol=1e-12, abs_tol=1e-12):
211
- raise TaskError("inconsistent hidden uncertainty")
212
- else:
213
- raise TaskError("unsupported marginal uncertainty model")
214
- item_map[item_id] = {"mean": mean, "sem": sem, "uncertainty_status": uncertainty_status}
215
- stored = hidden.get("pairs")
216
- if not isinstance(stored, list) or len(stored) != len(ids) * (len(ids) - 1) // 2:
217
- raise TaskError("invalid hidden pair count")
218
- stored_map = {}
219
- for pair in stored:
220
- if not isinstance(pair, dict):
221
- raise TaskError("invalid hidden pair")
222
- key = (pair.get("item_i"), pair.get("item_j"))
223
- if key in stored_map:
224
- raise TaskError("duplicate hidden pair")
225
- stored_map[key] = pair
226
- pairs = []
227
- for item_i, item_j in itertools.combinations(ids, 2):
228
- left, right = item_map[item_i], item_map[item_j]
229
- difference = left["mean"] - right["mean"]
230
- known = left["uncertainty_status"] == right["uncertainty_status"] == "known"
231
- se = difference_se(left["sem"], right["sem"], method) if known else None
232
- tolerance = k * se + epsilon if known else epsilon
233
- comparable = int(abs(difference) > tolerance)
234
- basis = "uncertainty_adjusted" if known else "reported_value"
235
- if not math.isfinite(difference) or not math.isfinite(tolerance) or (known and not math.isfinite(se)):
236
- raise TaskError("nonfinite pair arithmetic")
237
- pair = stored_map.get((item_i, item_j))
238
- if pair is None:
239
- raise TaskError("missing hidden pair")
240
- for name, expected, minimum in (("mean_difference", difference, None),):
241
- actual = finite(pair.get(name), name, minimum)
242
- if not math.isclose(actual, expected, rel_tol=1e-12, abs_tol=1e-10):
243
- raise TaskError("inconsistent hidden " + name)
244
- if known:
245
- if not math.isclose(finite(pair.get("difference_se"), "difference_se", 0), se,
246
- rel_tol=1e-12, abs_tol=1e-10):
247
- raise TaskError("inconsistent pair uncertainty")
248
- elif pair.get("difference_se") is not None:
249
- raise TaskError("unknown pair uncertainty must be null")
250
- if type(pair.get("comparable")) is not int or pair["comparable"] != comparable:
251
- raise TaskError("inconsistent comparable flag")
252
- if pair.get("uncertainty_status") != ("known" if known else "unknown"):
253
- raise TaskError("inconsistent pair uncertainty status")
254
- if pair.get("comparison_basis") != basis:
255
- raise TaskError("inconsistent comparison basis")
256
- pairs.append({"item_i": item_i, "item_j": item_j,
257
- "mean_difference": difference, "difference_se": se,
258
- "uncertainty_status": "known" if known else "unknown",
259
- "comparison_basis": basis, "comparable": comparable})
260
- total = sum(pair["comparable"] for pair in pairs)
261
- if total <= 0:
262
- raise TaskError("group has no scoreable pair under the declared rule")
263
- if type(hidden.get("comparable_pairs")) is not int or hidden["comparable_pairs"] != total:
264
- raise TaskError("inconsistent comparable pair count")
265
- validated.append({"group_id": gid, "item_ids": ids, "group_weight": group_weight,
266
- "pairs": pairs, "comparable_pairs": total})
267
- if set(hidden_map) != group_ids:
268
- raise TaskError("unknown hidden group")
269
- if not math.isclose(math.fsum(g["group_weight"] for g in validated), 1.0,
270
- rel_tol=0, abs_tol=1e-12):
271
- raise TaskError("group weights must sum to one")
272
- return validated
273
-
274
-
275
- def zero_report(contract, reason):
276
- return {"task_id": contract["task_id"], "task_version": contract["task_version"],
277
- "ranking_reward": 0.0, "submission_status": reason, "groups": [],
278
- "task_pair_accuracy": 0.0, "aggregation_zero_due_to_format": True,
279
- "official_reward": None, "review_status": "not_reviewed", "evaluation_status": "completed"}
280
-
281
-
282
- def grade_submission(contract, answers, submission, baseline):
283
- groups = validate_task(contract, answers)
284
- q = validate_baseline(contract, answers, baseline)
285
- if not isinstance(submission, dict):
286
- return zero_report(contract, "invalid_top_level_structure")
287
- if any(submission.get(key) != contract[key] for key in ("task_id", "task_version")):
288
- return zero_report(contract, "task_identity_mismatch")
289
- entries = submission.get("rankings")
290
- if not isinstance(entries, list):
291
- return zero_report(contract, "rankings_not_array")
292
- known = {g["group_id"] for g in groups}
293
- by_group = {}
294
- for entry in entries:
295
- if not isinstance(entry, dict) or not isinstance(entry.get("group_id"), str):
296
- return zero_report(contract, "unassignable_group_entry")
297
- if entry["group_id"] not in known:
298
- return zero_report(contract, "unknown_group_id")
299
- by_group.setdefault(entry["group_id"], []).append(entry)
300
- details = []
301
- for group in groups:
302
- gid = group["group_id"]
303
- matches = by_group.get(gid, [])
304
- item = {"group_id": gid, "group_weight": group["group_weight"],
305
- "group_pair_accuracy": 0.0, "status": "missing_group", "comparable_pairs": group["comparable_pairs"],
306
- "total_pairs": len(group["pairs"]), "item_count": len(group["item_ids"]),
307
- "coverage": group["comparable_pairs"] / len(group["pairs"])}
308
- if len(matches) > 1:
309
- item["status"] = "duplicate_group"
310
- elif len(matches) == 1:
311
- entry = matches[0]
312
- order = entry.get("order")
313
- if (not isinstance(order, list) or any(not isinstance(x, str) for x in order)
314
- or len(order) != len(group["item_ids"])
315
- or len(order) != len(set(order)) or set(order) != set(group["item_ids"])):
316
- item["status"] = "invalid_permutation"
317
- else:
318
- positions = {item_id: i for i, item_id in enumerate(order)}
319
- pair_details = []
320
- for pair in group["pairs"]:
321
- wrong = ((positions[pair["item_i"]] - positions[pair["item_j"]])
322
- * pair["mean_difference"]) < 0
323
- pair_details.append({**pair, "reversed": wrong,
324
- "counted_inversion": int(wrong and pair["comparable"] == 1)})
325
- lost = sum(pair["counted_inversion"] for pair in pair_details)
326
- total = group["comparable_pairs"]
327
- reward = 1.0 - lost / total
328
- item.update(status="valid", group_pair_accuracy=max(0.0, min(1.0, reward)),
329
- order=order, inversion_pairs=lost, correct_pairs=total-lost,
330
- comparable_pairs=total, total_pairs=len(pair_details),
331
- coverage=total/len(pair_details),
332
- reported_value_counted_pairs=sum(p["comparable"] for p in pair_details
333
- if p["comparison_basis"] == "reported_value"),
334
- uncertainty_adjusted_counted_pairs=sum(p["comparable"] for p in pair_details
335
- if p["comparison_basis"] == "uncertainty_adjusted"),
336
- all_reported_value_unequal_pairs=sum(p["mean_difference"] != 0 for p in pair_details),
337
- all_reported_value_correct_pairs=sum(p["mean_difference"] != 0 and not p["reversed"]
338
- for p in pair_details),
339
- pairs=pair_details)
340
- details.append(item)
341
- return {"task_id": contract["task_id"], "task_version": contract["task_version"],
342
- "task_pair_accuracy": aggregate_accuracy(details),
343
- "random_q95": q,
344
- "ranking_reward": reward_from_accuracy(aggregate_accuracy(details), q),
345
- "submission_status": "valid" if all(g["status"] == "valid" for g in details) else "group_errors",
346
- "groups": details, "official_reward": None,
347
- "review_status": "not_reviewed", "evaluation_status": "completed"}
348
-
349
-
350
- def read_regular(path, limit):
351
- try:
352
- fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
353
- try:
354
- info = os.fstat(fd)
355
- if not stat.S_ISREG(info.st_mode):
356
- raise SubmissionError("not_regular_file")
357
- if info.st_size > limit:
358
- raise SubmissionError("submission_too_large")
359
- with os.fdopen(fd, "rb", closefd=False) as stream:
360
- raw = stream.read(limit + 1)
361
- if len(raw) > limit:
362
- raise SubmissionError("submission_too_large")
363
- return raw
364
- finally:
365
- os.close(fd)
366
- except OSError as error:
367
- if error.errno in (errno.ENOENT, errno.ELOOP, errno.EISDIR, errno.ENOTDIR,
368
- errno.ENXIO, errno.EACCES, errno.EPERM):
369
- raise SubmissionError("missing_or_unreadable_submission") from error
370
- raise
371
-
372
-
373
- def grade_file(contract, answers, path, baseline):
374
- validate_task(contract, answers)
375
- q = validate_baseline(contract, answers, baseline)
376
- try:
377
- raw = read_regular(path, contract["max_submission_bytes"])
378
- submission = load_json(raw)
379
- except (SubmissionError, UnicodeError, json.JSONDecodeError, RecursionError) as error:
380
- report = zero_report(contract, str(error) or type(error).__name__)
381
- report["random_q95"] = q
382
- return report
383
- report = grade_submission(contract, answers, submission, baseline)
384
- report["random_q95"] = q
385
- report["submission_sha256"] = hashlib.sha256(raw).hexdigest()
386
- return report
387
-
388
-
389
- def dump(path, value):
390
- path.write_text(json.dumps(value, indent=2, allow_nan=False) + "\n", encoding="utf-8")
391
-
392
-
393
- def main():
394
- parser = argparse.ArgumentParser()
395
- parser.add_argument("--contract", type=Path, default=Path("/tests/contract.json"))
396
- parser.add_argument("--answers", type=Path, default=Path("/tests/answers.json"))
397
- parser.add_argument("--baseline", type=Path, default=Path("/tests/random_baseline.json"))
398
- parser.add_argument("--submission", type=Path, default=Path("/logs/artifacts/rankings.json"))
399
- parser.add_argument("--output-dir", type=Path, default=Path("/logs/verifier"))
400
- args = parser.parse_args()
401
- args.output_dir.mkdir(parents=True, exist_ok=True)
402
- for name in ("reward.txt", "reward.json", "report.json", "evaluator_error.json"):
403
- (args.output_dir / name).unlink(missing_ok=True)
404
- try:
405
- contract = load_json(args.contract.read_bytes())
406
- answers = load_json(args.answers.read_bytes())
407
- baseline = load_json(args.baseline.read_bytes())
408
- report = grade_file(contract, answers, args.submission, baseline)
409
- except (TaskError, SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
410
- dump(args.output_dir / "evaluator_error.json", {"status": "task_error", "error": str(error)})
411
- raise SystemExit(2) from error
412
- dump(args.output_dir / "report.json", report)
413
- (args.output_dir / "reward.txt").write_text(str(report["ranking_reward"]) + "\n", encoding="utf-8")
414
- print(json.dumps({"ranking_reward": report["ranking_reward"],
415
- "submission_status": report["submission_status"],
416
- "task_pair_accuracy": report["task_pair_accuracy"],
417
- "random_q95": report["random_q95"],
418
- "aggregation_zero_due_to_format": report.get("aggregation_zero_due_to_format", False),
419
- "official_reward": None, "review_status": "not_reviewed", "evaluation_status": "completed"}))
420
-
421
-
422
- if __name__ == "__main__":
423
- main()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tasks/alkene-transfer-conditions-ranking/tests/random_baseline.json DELETED
@@ -1,67 +0,0 @@
1
- {
2
- "method": "uniform_complete_permutations",
3
- "generator": "numpy.random.PCG64",
4
- "numpy_version": "2.4.4",
5
- "quantile": 0.95,
6
- "quantile_method": "higher",
7
- "calibration_count": 200000,
8
- "validation_count": 200000,
9
- "calibration_seed": 6626574861231631612,
10
- "validation_seed": 7652489934321606869,
11
- "q": 0.769047619047619,
12
- "task_id": "alkene-transfer-conditions-ranking",
13
- "task_version": "1.0.0",
14
- "group_order": [
15
- "g1",
16
- "g2",
17
- "g3"
18
- ],
19
- "item_order": {
20
- "g1": [
21
- "exp_41bd98dba742",
22
- "exp_484d7efaebd5",
23
- "exp_389c830f7e89",
24
- "exp_dd127eb2b1f5",
25
- "exp_8545262bd121",
26
- "exp_369ffbca2e28",
27
- "exp_cac98fcd26c1"
28
- ],
29
- "g2": [
30
- "exp_066e33c896b9",
31
- "exp_fba92a14e6bb",
32
- "exp_9279aa7cd389",
33
- "exp_8e5983854cd1",
34
- "exp_343b2ac1c9a7",
35
- "exp_e01ea673fa5d",
36
- "exp_ea300526f709"
37
- ],
38
- "g3": [
39
- "exp_529d2754aa6a",
40
- "exp_94c775c85236"
41
- ]
42
- },
43
- "calibration_code_sha256": "cbb437c19ae80d85f28374fe62c71e8052975ddc48c447fe89b9fb9913e8c6b1",
44
- "relations_sha256": "e7005ac4aac641a5d46a926a131fdf6b4a1af06321d6a38efaa076527a3e6e5d",
45
- "weights_sha256": "2679d5eb44e50e29d60ff733779829e26a835b7720da8cd85a869493156ce9ea",
46
- "validation": {
47
- "task_pair_accuracy": {
48
- "mean": 0.49968663888888887,
49
- "mean_standard_error": 0.00040975817618166564
50
- },
51
- "task_spearman": {
52
- "mean": -0.0006088104998788662,
53
- "mean_standard_error": 0.0008605907829932248
54
- },
55
- "ranking_reward": {
56
- "mean": 0.006584793814432992,
57
- "mean_standard_error": 8.896824877574369e-05
58
- },
59
- "zero_reward_fraction": 0.949825
60
- },
61
- "bindings": {
62
- "contract_sha256": "b1f08c9ac5f8d1da9bfd60d746c6fdad40282ffe3663c6d61a5d3eb704b4df03",
63
- "answers_sha256": "c47ed0d746703529048a915a8383bf0b4fb3cd60e042599699b69c47466f6f28",
64
- "grader_sha256": "82c9d3b9c1fa5c2e2a39852e315df9e04714dee2ffa9eb64373014d8b4aeb381",
65
- "auxiliary_grader_sha256": "b91d5af76c52dca0cb7195777a0cbea690a91af15f6bdbb0b13714e03ab45492"
66
- }
67
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tasks/alkene-transfer-conditions-ranking/tests/scoring_hashes.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "answers.json": "13b6b71e945ced4ff3f2239aff516c608343e63ce09063c34dec88a11c5ad29c",
3
+ "contract.json": "907b76dd30abc155c7ea34338f09fe92d5622e153a29fbc0363f6ed133dc014f",
4
+ "grade.py": "0c25490ea9639fd4e73fde4741f4b4e02bdbd42962779348694e50e33a121550"
5
+ }
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/contract.json CHANGED
@@ -1,33 +1,27 @@
1
  {
2
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "reaction_settings",
7
  "label": "Reaction settings",
8
  "item_ids": [
9
- "exp_2dfe2f22dd32",
10
- "exp_34e91dee44f6",
11
  "exp_0003b83bcce3",
12
- "exp_595fa6ce8e2c",
13
  "exp_1da97213829d",
14
- "exp_2da4d7cbbf59",
15
  "exp_2191e72320e5",
16
- "exp_1638cbaa31ee",
17
- "exp_3e426daf35ae"
 
 
18
  ],
19
  "direction": "ascending",
20
  "group_weight": 1.0,
21
- "k": 2,
22
- "epsilon": 0,
23
- "uncertainty_model": "optional_reported_uncertainty",
24
- "difference_standard_error": "sem_sum_upper_bound",
25
  "endpoint": {
26
  "name": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard.",
27
  "unit": "%",
28
  "definition_en": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
29
- },
30
- "uncertainty_definition": "Independent and technical replicate counts and reliable difference uncertainties are not reported for this panel. All errors remain unknown; the reported-value branch is used."
31
  }
32
  ],
33
  "scoring_method": "spearman",
@@ -39,18 +33,9 @@
39
  "spearman_tie_method": "average",
40
  "spearman_undefined_policy": "task_error",
41
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
42
- "secondary_scoring_method": "equal_pair_inversion_random95",
43
- "ranking_reward_range": [
44
- 0,
45
- 1
46
- ],
47
- "random_quantile": 0.95,
48
- "unknown_uncertainty_policy": "use_reported_value",
49
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
50
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
51
  "format_errors": {
52
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
53
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
54
  },
55
  "max_submission_bytes": 1048576,
56
  "minimum_conditions_per_group": 2,
@@ -68,6 +53,13 @@
68
  "sources_path": "/logs/artifacts/sources.json",
69
  "analysis_path": "/logs/artifacts/analysis/",
70
  "forbidden_result_scope": "Same-type forbidden results are nickel-mediated asymmetric difluoromethylations of unactivated alkyl halides with zinc difluoromethyl reagents, including yields, conversion, enantiomeric excess, additive/ligand/nickel/oxidant/solvent comparisons and outcome-fitted predictors.",
71
- "pair_boundary_arithmetic": "exact_decimal",
72
- "pair_boundary_definition": "Interpret each stored numeric input by its decimal text and evaluate the strict pair inequality exactly. Equality is not a decidable pair. For independent SEM quadrature, compare positive squared sides exactly; no floating-point tolerance is added to k or epsilon."
 
 
 
 
 
 
 
73
  }
 
1
  {
2
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "reaction_settings",
7
  "label": "Reaction settings",
8
  "item_ids": [
 
 
9
  "exp_0003b83bcce3",
10
+ "exp_1638cbaa31ee",
11
  "exp_1da97213829d",
 
12
  "exp_2191e72320e5",
13
+ "exp_2dfe2f22dd32",
14
+ "exp_34e91dee44f6",
15
+ "exp_3e426daf35ae",
16
+ "exp_595fa6ce8e2c"
17
  ],
18
  "direction": "ascending",
19
  "group_weight": 1.0,
 
 
 
 
20
  "endpoint": {
21
  "name": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard.",
22
  "unit": "%",
23
  "definition_en": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
24
+ }
 
25
  }
26
  ],
27
  "scoring_method": "spearman",
 
33
  "spearman_tie_method": "average",
34
  "spearman_undefined_policy": "task_error",
35
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
36
  "format_errors": {
37
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
38
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
39
  },
40
  "max_submission_bytes": 1048576,
41
  "minimum_conditions_per_group": 2,
 
53
  "sources_path": "/logs/artifacts/sources.json",
54
  "analysis_path": "/logs/artifacts/analysis/",
55
  "forbidden_result_scope": "Same-type forbidden results are nickel-mediated asymmetric difluoromethylations of unactivated alkyl halides with zinc difluoromethyl reagents, including yields, conversion, enantiomeric excess, additive/ligand/nickel/oxidant/solvent comparisons and outcome-fitted predictors.",
56
+ "supporting_artifact_review": {
57
+ "policy_stop_marker": "SPU_POLICY_STOP",
58
+ "required_files": [
59
+ "report.md",
60
+ "sources.json"
61
+ ],
62
+ "required_directory": "analysis",
63
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
64
+ }
65
  }
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/experiment.md CHANGED
@@ -1,6 +1,6 @@
1
  # Experimental definition
2
 
3
- N-(3,5-Dimethoxyphenyl)-3-iodopentanamide is coupled with a zinc difluoromethyl reagent using the drawn chiral nickel/ligand system and diphenyliodonium triflate. Predict the product 19F NMR yield across the nine additive identities.
4
 
5
  ## Common conditions
6
 
@@ -34,8 +34,8 @@ N-(3,5-Dimethoxyphenyl)-3-iodopentanamide is coupled with a zinc difluoromethyl
34
  | cooling_before_injection_time_min | `10` |
35
  | cooling_before_injection_temperature_C | `-10` |
36
  | addition_order | `Ni precursor and ligand in diglyme; premix; additive and iodonium salt; cool; inject zinc reagent and substrate in THF` |
37
- | thf_preparation | `anhydrous THF dried over4 A molecular sieves and stored in glovebox` |
38
- | workup | `water quench,three diethyl ether extractions,brine wash,dry over magnesium sulfate and concentrate; silica chromatography follows for isolation` |
39
  | nominal_total_solvent_charge_mL | `0.5` |
40
  | diglyme_charge_mL | `0.4` |
41
  | thf_injection_charge_mL | `0.1` |
@@ -49,7 +49,7 @@ In a nitrogen glovebox, premix nickel bromide dimethoxyethane complex and the dr
49
 
50
  ## Item settings
51
 
52
- Expand each item in targets.csv or protocol.json together with the common conditions. Every selected item is listed using the same fields.
53
 
54
  ## Source limitations relevant to the definition
55
 
 
1
  # Experimental definition
2
 
3
+ N-(3,5-Dimethoxyphenyl)-3-iodopentanamide is coupled with a zinc difluoromethyl reagent using the drawn chiral nickel/ligand system and diphenyliodonium triflate. Predict the product 19F NMR yield across the eight additive identities.
4
 
5
  ## Common conditions
6
 
 
34
  | cooling_before_injection_time_min | `10` |
35
  | cooling_before_injection_temperature_C | `-10` |
36
  | addition_order | `Ni precursor and ligand in diglyme; premix; additive and iodonium salt; cool; inject zinc reagent and substrate in THF` |
37
+ | thf_preparation | `anhydrous THF dried over 4 A molecular sieves and stored in glovebox` |
38
+ | workup | `water quench, three diethyl ether extractions, brine wash, dry over magnesium sulfate and concentrate; silica chromatography follows for isolation` |
39
  | nominal_total_solvent_charge_mL | `0.5` |
40
  | diglyme_charge_mL | `0.4` |
41
  | thf_injection_charge_mL | `0.1` |
 
49
 
50
  ## Item settings
51
 
52
+ Expand each item in targets.csv or protocol.json together with the common conditions.
53
 
54
  ## Source limitations relevant to the definition
55
 
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/information_policy.md CHANGED
@@ -11,3 +11,5 @@ Forbidden examples:
11
  - Measured yields, conversions, selectivities, equivalent outcome proxies or condition rankings inside the declared reaction scope, and predictors fitted to those outcomes.
12
 
13
  Use audited tools for all searches, downloads and code-initiated network requests. Check both requests and returned content before use. Do not circumvent access controls or ask another model/service for forbidden results. If a permitted query unexpectedly exposes prohibited external results, immediately stop prediction and further search; record only the triggering step, source and stopping reason in report.md without restating results. Do not retrieve or reconstruct specific experimental results from your own model memory; briefly disclose a noticed recollection in your report and continue using permitted inputs and reasoning.
 
 
 
11
  - Measured yields, conversions, selectivities, equivalent outcome proxies or condition rankings inside the declared reaction scope, and predictors fitted to those outcomes.
12
 
13
  Use audited tools for all searches, downloads and code-initiated network requests. Check both requests and returned content before use. Do not circumvent access controls or ask another model/service for forbidden results. If a permitted query unexpectedly exposes prohibited external results, immediately stop prediction and further search; record only the triggering step, source and stopping reason in report.md without restating results. Do not retrieve or reconstruct specific experimental results from your own model memory; briefly disclose a noticed recollection in your report and continue using permitted inputs and reasoning.
14
+
15
+ For a stop under the information policy, put `SPU_POLICY_STOP` on the first line of `report.md` and record the triggering step, source and reason below it. Withdraw any provisional ranking.
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/protocol.json CHANGED
@@ -28,8 +28,8 @@
28
  "cooling_before_injection_time_min": 10,
29
  "cooling_before_injection_temperature_C": -10,
30
  "addition_order": "Ni precursor and ligand in diglyme; premix; additive and iodonium salt; cool; inject zinc reagent and substrate in THF",
31
- "thf_preparation": "anhydrous THF dried over4 A molecular sieves and stored in glovebox",
32
- "workup": "water quench,three diethyl ether extractions,brine wash,dry over magnesium sulfate and concentrate; silica chromatography follows for isolation",
33
  "nominal_total_solvent_charge_mL": 0.5,
34
  "diglyme_charge_mL": 0.4,
35
  "thf_injection_charge_mL": 0.1,
@@ -54,32 +54,29 @@
54
  "measurement_rule": "Use measurements_by_group for each group. The measurement field is a default summary and does not override a group-specific endpoint.",
55
  "inheritance_rule": "common_conditions apply unless the same field is explicitly overridden by the item; item settings are literal experimental facts, not predicted outcomes.",
56
  "items": {
57
- "exp_2dfe2f22dd32": {
58
- "additive": "magnesium acetate"
59
- },
60
- "exp_34e91dee44f6": {
61
- "additive": "triethylamine"
62
- },
63
  "exp_0003b83bcce3": {
64
  "additive": "trimethylsilyl chloride"
65
  },
66
- "exp_595fa6ce8e2c": {
67
- "additive": "tribenzylamine"
68
  },
69
  "exp_1da97213829d": {
70
  "additive": "zinc acetate"
71
  },
72
- "exp_2da4d7cbbf59": {
73
- "additive": "triphenylamine"
74
- },
75
  "exp_2191e72320e5": {
76
  "additive": "sodium carbonate"
77
  },
78
- "exp_1638cbaa31ee": {
79
- "additive": "diisopropylamine"
 
 
 
80
  },
81
  "exp_3e426daf35ae": {
82
  "additive": "N,N-diisopropylethylamine"
 
 
 
83
  }
84
  }
85
  }
 
28
  "cooling_before_injection_time_min": 10,
29
  "cooling_before_injection_temperature_C": -10,
30
  "addition_order": "Ni precursor and ligand in diglyme; premix; additive and iodonium salt; cool; inject zinc reagent and substrate in THF",
31
+ "thf_preparation": "anhydrous THF dried over 4 A molecular sieves and stored in glovebox",
32
+ "workup": "water quench, three diethyl ether extractions, brine wash, dry over magnesium sulfate and concentrate; silica chromatography follows for isolation",
33
  "nominal_total_solvent_charge_mL": 0.5,
34
  "diglyme_charge_mL": 0.4,
35
  "thf_injection_charge_mL": 0.1,
 
54
  "measurement_rule": "Use measurements_by_group for each group. The measurement field is a default summary and does not override a group-specific endpoint.",
55
  "inheritance_rule": "common_conditions apply unless the same field is explicitly overridden by the item; item settings are literal experimental facts, not predicted outcomes.",
56
  "items": {
 
 
 
 
 
 
57
  "exp_0003b83bcce3": {
58
  "additive": "trimethylsilyl chloride"
59
  },
60
+ "exp_1638cbaa31ee": {
61
+ "additive": "diisopropylamine"
62
  },
63
  "exp_1da97213829d": {
64
  "additive": "zinc acetate"
65
  },
 
 
 
66
  "exp_2191e72320e5": {
67
  "additive": "sodium carbonate"
68
  },
69
+ "exp_2dfe2f22dd32": {
70
+ "additive": "magnesium acetate"
71
+ },
72
+ "exp_34e91dee44f6": {
73
+ "additive": "triethylamine"
74
  },
75
  "exp_3e426daf35ae": {
76
  "additive": "N,N-diisopropylethylamine"
77
+ },
78
+ "exp_595fa6ce8e2c": {
79
+ "additive": "tribenzylamine"
80
  }
81
  }
82
  }
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/environment/public/targets.csv CHANGED
@@ -1,10 +1,9 @@
1
  group_id,item_id,settings_ref,additive
2
- reaction_settings,exp_2dfe2f22dd32,protocol.json#exp_2dfe2f22dd32,magnesium acetate
3
- reaction_settings,exp_34e91dee44f6,protocol.json#exp_34e91dee44f6,triethylamine
4
  reaction_settings,exp_0003b83bcce3,protocol.json#exp_0003b83bcce3,trimethylsilyl chloride
5
- reaction_settings,exp_595fa6ce8e2c,protocol.json#exp_595fa6ce8e2c,tribenzylamine
6
  reaction_settings,exp_1da97213829d,protocol.json#exp_1da97213829d,zinc acetate
7
- reaction_settings,exp_2da4d7cbbf59,protocol.json#exp_2da4d7cbbf59,triphenylamine
8
  reaction_settings,exp_2191e72320e5,protocol.json#exp_2191e72320e5,sodium carbonate
9
- reaction_settings,exp_1638cbaa31ee,protocol.json#exp_1638cbaa31ee,diisopropylamine
 
10
  reaction_settings,exp_3e426daf35ae,protocol.json#exp_3e426daf35ae,"N,N-diisopropylethylamine"
 
 
1
  group_id,item_id,settings_ref,additive
 
 
2
  reaction_settings,exp_0003b83bcce3,protocol.json#exp_0003b83bcce3,trimethylsilyl chloride
3
+ reaction_settings,exp_1638cbaa31ee,protocol.json#exp_1638cbaa31ee,diisopropylamine
4
  reaction_settings,exp_1da97213829d,protocol.json#exp_1da97213829d,zinc acetate
 
5
  reaction_settings,exp_2191e72320e5,protocol.json#exp_2191e72320e5,sodium carbonate
6
+ reaction_settings,exp_2dfe2f22dd32,protocol.json#exp_2dfe2f22dd32,magnesium acetate
7
+ reaction_settings,exp_34e91dee44f6,protocol.json#exp_34e91dee44f6,triethylamine
8
  reaction_settings,exp_3e426daf35ae,protocol.json#exp_3e426daf35ae,"N,N-diisopropylethylamine"
9
+ reaction_settings,exp_595fa6ce8e2c,protocol.json#exp_595fa6ce8e2c,tribenzylamine
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/instruction.md CHANGED
@@ -14,7 +14,7 @@ You have a strict time limit of 8 hours (28,800 seconds). Submit the required fi
14
 
15
  ## Prediction question
16
 
17
- N-(3,5-Dimethoxyphenyl)-3-iodopentanamide is coupled with a zinc difluoromethyl reagent using the drawn chiral nickel/ligand system and diphenyliodonium triflate. Predict the product 19F NMR yield across the nine additive identities.
18
 
19
  Rank each comparison group from lowest to highest using that group's specified endpoint and unit. Submit rankings only; absolute values are not required.
20
 
@@ -30,7 +30,7 @@ Outcome definitions by group:
30
 
31
  ## Comparison groups
32
 
33
- - `reaction_settings`: Reaction settings; 9 items; weight 1.0; ascending 19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard. (%).
34
 
35
  ## Allowed and prohibited information
36
 
@@ -52,71 +52,61 @@ You have 8 CPU cores, 16,384 MiB system RAM and one NVIDIA H20 GPU, shared by al
52
 
53
  ## Submission
54
 
55
- Write UTF-8 JSON to `/logs/artifacts/rankings.json` (at most 1 MiB, unique object keys). Include the exact task identity and exactly one entry per group; each order must contain every assigned item once, from lowest to highest. The following is a format example in public display order, not a predicted answer:
56
 
57
  ```json
58
  {
59
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
60
- "task_version": "1.0.0",
61
  "rankings": [
62
  {
63
  "group_id": "reaction_settings",
64
  "order": [
65
- "exp_2dfe2f22dd32",
66
- "exp_34e91dee44f6",
67
  "exp_0003b83bcce3",
68
- "exp_595fa6ce8e2c",
69
  "exp_1da97213829d",
70
- "exp_2da4d7cbbf59",
71
  "exp_2191e72320e5",
72
- "exp_1638cbaa31ee",
73
- "exp_3e426daf35ae"
 
 
74
  ]
75
  }
76
  ]
77
  }
78
  ```
79
 
80
- Write an English account of scientific reasoning, calculations, assumptions and uncertainty to `/logs/artifacts/report.md`; source records to `/logs/artifacts/sources.json`; supporting scripts and calculations to `/logs/artifacts/analysis/`. The evaluator uses the last submission received before the deadline. Empty or missing submissions receive core -1 and auxiliary 0.
81
 
82
- ## Ranking scores
83
 
84
- The primary score is standard Spearman rank correlation, from -1 to 1, computed separately in each group and combined with fixed group weights. Equal measured values receive average true ranks; predictions remain strict permutations. Correct/reverse rankings score 1/-1 when true values have no ties; with true ties the best strict ranking can score less than 1. Random complete rankings have expected correlation 0.
85
 
86
- The auxiliary score counts each decidable pair equally within its group, combines group accuracies using the same weights, then applies one whole-task random 95th-percentile threshold. Known uncertainty uses the declared difference-error rule; unknown uncertainty uses reported values with epsilon and does not exclude unequal results. The threshold is hidden. Core and auxiliary scores are recorded separately; neither meaning of zero proves the absence of predictive ability. Detailed algorithms and all format penalties are in the following authoritative public contract, from which the companion files are generated.
87
-
88
- <!-- public-contract -->
89
  ```json
90
  {
91
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
92
- "task_version": "1.0.0",
93
  "groups": [
94
  {
95
  "group_id": "reaction_settings",
96
  "label": "Reaction settings",
97
  "item_ids": [
98
- "exp_2dfe2f22dd32",
99
- "exp_34e91dee44f6",
100
  "exp_0003b83bcce3",
101
- "exp_595fa6ce8e2c",
102
  "exp_1da97213829d",
103
- "exp_2da4d7cbbf59",
104
  "exp_2191e72320e5",
105
- "exp_1638cbaa31ee",
106
- "exp_3e426daf35ae"
 
 
107
  ],
108
  "direction": "ascending",
109
  "group_weight": 1.0,
110
- "k": 2,
111
- "epsilon": 0,
112
- "uncertainty_model": "optional_reported_uncertainty",
113
- "difference_standard_error": "sem_sum_upper_bound",
114
  "endpoint": {
115
  "name": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard.",
116
  "unit": "%",
117
  "definition_en": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
118
- },
119
- "uncertainty_definition": "Independent and technical replicate counts and reliable difference uncertainties are not reported for this panel. All errors remain unknown; the reported-value branch is used."
120
  }
121
  ],
122
  "scoring_method": "spearman",
@@ -128,18 +118,9 @@ The auxiliary score counts each decidable pair equally within its group, combine
128
  "spearman_tie_method": "average",
129
  "spearman_undefined_policy": "task_error",
130
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
131
- "secondary_scoring_method": "equal_pair_inversion_random95",
132
- "ranking_reward_range": [
133
- 0,
134
- 1
135
- ],
136
- "random_quantile": 0.95,
137
- "unknown_uncertainty_policy": "use_reported_value",
138
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
139
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
140
  "format_errors": {
141
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
142
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
143
  },
144
  "max_submission_bytes": 1048576,
145
  "minimum_conditions_per_group": 2,
@@ -157,8 +138,15 @@ The auxiliary score counts each decidable pair equally within its group, combine
157
  "sources_path": "/logs/artifacts/sources.json",
158
  "analysis_path": "/logs/artifacts/analysis/",
159
  "forbidden_result_scope": "Same-type forbidden results are nickel-mediated asymmetric difluoromethylations of unactivated alkyl halides with zinc difluoromethyl reagents, including yields, conversion, enantiomeric excess, additive/ligand/nickel/oxidant/solvent comparisons and outcome-fitted predictors.",
160
- "pair_boundary_arithmetic": "exact_decimal",
161
- "pair_boundary_definition": "Interpret each stored numeric input by its decimal text and evaluate the strict pair inequality exactly. Equality is not a decidable pair. For independent SEM quadrature, compare positive squared sides exactly; no floating-point tolerance is added to k or epsilon."
 
 
 
 
 
 
 
162
  }
163
  ```
164
  <!-- /public-contract -->
 
14
 
15
  ## Prediction question
16
 
17
+ N-(3,5-Dimethoxyphenyl)-3-iodopentanamide is coupled with a zinc difluoromethyl reagent using the drawn chiral nickel/ligand system and diphenyliodonium triflate. Predict the product 19F NMR yield across the eight additive identities.
18
 
19
  Rank each comparison group from lowest to highest using that group's specified endpoint and unit. Submit rankings only; absolute values are not required.
20
 
 
30
 
31
  ## Comparison groups
32
 
33
+ - `reaction_settings`: Reaction settings; 8 items; weight 1.0; ascending 19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard. (%).
34
 
35
  ## Allowed and prohibited information
36
 
 
52
 
53
  ## Submission
54
 
55
+ Write UTF-8 JSON to `/logs/artifacts/rankings.json` (at most 1 MiB, unique object keys). Include the exact task identity and exactly one entry per group; each order must contain every assigned item once, from lowest to highest. The example illustrates the submission format. Submit your predicted order for each group.
56
 
57
  ```json
58
  {
59
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
60
+ "task_version": "1.1.0",
61
  "rankings": [
62
  {
63
  "group_id": "reaction_settings",
64
  "order": [
 
 
65
  "exp_0003b83bcce3",
66
+ "exp_1638cbaa31ee",
67
  "exp_1da97213829d",
 
68
  "exp_2191e72320e5",
69
+ "exp_2dfe2f22dd32",
70
+ "exp_34e91dee44f6",
71
+ "exp_3e426daf35ae",
72
+ "exp_595fa6ce8e2c"
73
  ]
74
  }
75
  ]
76
  }
77
  ```
78
 
79
+ Write an English account of scientific reasoning, calculations, assumptions and uncertainty to `/logs/artifacts/report.md`; source records to `/logs/artifacts/sources.json`; supporting scripts and calculations to `/logs/artifacts/analysis/`. Only rankings.json contributes to the Spearman reward. The verifier records the presence of the supporting artifacts for review. For a stop under the information policy, put `SPU_POLICY_STOP` on the first line of `report.md` and record the triggering step, source and reason below it. Withdraw any provisional ranking. The evaluator uses the last submission received before the deadline. Empty or missing submissions receive -1.
80
 
81
+ ## Ranking score
82
 
83
+ The task score is the weighted mean of the group Spearman rank correlations and ranges from -1 to 1. Equal measurements share their average rank. Submit a strict ordering for each group. When measurements are distinct, the measured order scores 1 and its reverse scores -1. Tied measurements can place the highest attainable correlation below 1. A uniformly random ordering has expected correlation 0.
84
 
 
 
 
85
  ```json
86
  {
87
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
88
+ "task_version": "1.1.0",
89
  "groups": [
90
  {
91
  "group_id": "reaction_settings",
92
  "label": "Reaction settings",
93
  "item_ids": [
 
 
94
  "exp_0003b83bcce3",
95
+ "exp_1638cbaa31ee",
96
  "exp_1da97213829d",
 
97
  "exp_2191e72320e5",
98
+ "exp_2dfe2f22dd32",
99
+ "exp_34e91dee44f6",
100
+ "exp_3e426daf35ae",
101
+ "exp_595fa6ce8e2c"
102
  ],
103
  "direction": "ascending",
104
  "group_weight": 1.0,
 
 
 
 
105
  "endpoint": {
106
  "name": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard.",
107
  "unit": "%",
108
  "definition_en": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
109
+ }
 
110
  }
111
  ],
112
  "scoring_method": "spearman",
 
118
  "spearman_tie_method": "average",
119
  "spearman_undefined_policy": "task_error",
120
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
121
  "format_errors": {
122
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
123
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
124
  },
125
  "max_submission_bytes": 1048576,
126
  "minimum_conditions_per_group": 2,
 
138
  "sources_path": "/logs/artifacts/sources.json",
139
  "analysis_path": "/logs/artifacts/analysis/",
140
  "forbidden_result_scope": "Same-type forbidden results are nickel-mediated asymmetric difluoromethylations of unactivated alkyl halides with zinc difluoromethyl reagents, including yields, conversion, enantiomeric excess, additive/ligand/nickel/oxidant/solvent comparisons and outcome-fitted predictors.",
141
+ "supporting_artifact_review": {
142
+ "policy_stop_marker": "SPU_POLICY_STOP",
143
+ "required_files": [
144
+ "report.md",
145
+ "sources.json"
146
+ ],
147
+ "required_directory": "analysis",
148
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
149
+ }
150
  }
151
  ```
152
  <!-- /public-contract -->
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/solution/solve.sh CHANGED
@@ -1,10 +1,10 @@
1
  #!/bin/sh
2
  set -eu
3
  mkdir -p /logs/artifacts
4
- cat > /logs/artifacts/rankings.json <<'PROJECTAA_ORACLE'
5
  {
6
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
7
- "task_version": "1.0.0",
8
  "rankings": [
9
  {
10
  "group_id": "reaction_settings",
@@ -12,7 +12,6 @@ cat > /logs/artifacts/rankings.json <<'PROJECTAA_ORACLE'
12
  "exp_0003b83bcce3",
13
  "exp_595fa6ce8e2c",
14
  "exp_1da97213829d",
15
- "exp_2da4d7cbbf59",
16
  "exp_2dfe2f22dd32",
17
  "exp_34e91dee44f6",
18
  "exp_2191e72320e5",
@@ -22,4 +21,4 @@ cat > /logs/artifacts/rankings.json <<'PROJECTAA_ORACLE'
22
  }
23
  ]
24
  }
25
- PROJECTAA_ORACLE
 
1
  #!/bin/sh
2
  set -eu
3
  mkdir -p /logs/artifacts
4
+ cat > /logs/artifacts/rankings.json <<'EOF'
5
  {
6
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
7
+ "task_version": "1.1.0",
8
  "rankings": [
9
  {
10
  "group_id": "reaction_settings",
 
12
  "exp_0003b83bcce3",
13
  "exp_595fa6ce8e2c",
14
  "exp_1da97213829d",
 
15
  "exp_2dfe2f22dd32",
16
  "exp_34e91dee44f6",
17
  "exp_2191e72320e5",
 
21
  }
22
  ]
23
  }
24
+ EOF
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/task.toml CHANGED
@@ -1,9 +1,7 @@
1
- schema_version = "1.2"
2
  [metadata]
3
  task_id = "alkyl-iodide-nickel-difluoromethylation-additive-yield"
4
- task_version = "1.0.0"
5
- pipeline_version = "19"
6
- scientist_prompt_version = "projectaa-scientist-en-v7-ranking"
7
  [agent]
8
  user = "solver"
9
  timeout_sec = 28800.0
@@ -14,7 +12,7 @@ memory_mb = 16384
14
  storage_mb = 20480
15
  gpus = 1
16
  gpu_types = ["H20"]
17
- allow_internet = true
18
  [verifier]
19
  timeout_sec = 120.0
20
  environment_mode = "separate"
@@ -24,4 +22,4 @@ cpus = 1
24
  memory_mb = 1024
25
  storage_mb = 2048
26
  gpus = 0
27
- allow_internet = false
 
1
+ schema_version = "1.4"
2
  [metadata]
3
  task_id = "alkyl-iodide-nickel-difluoromethylation-additive-yield"
4
+ task_version = "1.1.0"
 
 
5
  [agent]
6
  user = "solver"
7
  timeout_sec = 28800.0
 
12
  storage_mb = 20480
13
  gpus = 1
14
  gpu_types = ["H20"]
15
+ network_mode = "public"
16
  [verifier]
17
  timeout_sec = 120.0
18
  environment_mode = "separate"
 
22
  memory_mb = 1024
23
  storage_mb = 2048
24
  gpus = 0
25
+ network_mode = "no-network"
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/answers.json CHANGED
@@ -1,465 +1,67 @@
1
  {
2
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "reaction_settings",
7
  "items": [
8
- {
9
- "item_id": "exp_2dfe2f22dd32",
10
- "mean": 59,
11
- "n": null,
12
- "uncertainty_status": "unknown",
13
- "uncertainty_scale": null,
14
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
15
- "quality_status": "retained_with_flag",
16
- "warnings": [
17
- "Replication and difference uncertainty are not reported."
18
- ],
19
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry6",
20
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
21
- },
22
- {
23
- "item_id": "exp_34e91dee44f6",
24
- "mean": 60,
25
- "n": null,
26
- "uncertainty_status": "unknown",
27
- "uncertainty_scale": null,
28
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
29
- "quality_status": "retained_with_flag",
30
- "warnings": [
31
- "Replication and difference uncertainty are not reported."
32
- ],
33
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry1",
34
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
35
- },
36
  {
37
  "item_id": "exp_0003b83bcce3",
38
  "mean": 40,
39
  "n": null,
40
  "uncertainty_status": "unknown",
41
- "uncertainty_scale": null,
42
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
43
- "quality_status": "retained_with_flag",
44
- "warnings": [
45
- "Replication and difference uncertainty are not reported."
46
- ],
47
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry8",
48
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
49
  },
50
  {
51
- "item_id": "exp_595fa6ce8e2c",
52
- "mean": 46,
53
  "n": null,
54
  "uncertainty_status": "unknown",
55
- "uncertainty_scale": null,
56
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
57
- "quality_status": "retained_with_flag",
58
- "warnings": [
59
- "Replication and difference uncertainty are not reported."
60
- ],
61
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry4",
62
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
63
  },
64
  {
65
  "item_id": "exp_1da97213829d",
66
  "mean": 58,
67
  "n": null,
68
  "uncertainty_status": "unknown",
69
- "uncertainty_scale": null,
70
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
71
- "quality_status": "retained_with_flag",
72
- "warnings": [
73
- "Replication and difference uncertainty are not reported."
74
- ],
75
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry9",
76
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
77
  },
78
  {
79
- "item_id": "exp_2da4d7cbbf59",
80
- "mean": 59,
81
  "n": null,
82
  "uncertainty_status": "unknown",
83
- "uncertainty_scale": null,
84
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
85
- "quality_status": "retained_with_flag",
86
- "warnings": [
87
- "Replication and difference uncertainty are not reported."
88
- ],
89
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry5",
90
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
91
  },
92
  {
93
- "item_id": "exp_2191e72320e5",
94
- "mean": 61,
95
  "n": null,
96
  "uncertainty_status": "unknown",
97
- "uncertainty_scale": null,
98
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
99
- "quality_status": "retained_with_flag",
100
- "warnings": [
101
- "Replication and difference uncertainty are not reported."
102
- ],
103
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry7",
104
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
105
  },
106
  {
107
- "item_id": "exp_1638cbaa31ee",
108
- "mean": 62,
109
  "n": null,
110
  "uncertainty_status": "unknown",
111
- "uncertainty_scale": null,
112
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
113
- "quality_status": "retained_with_flag",
114
- "warnings": [
115
- "Replication and difference uncertainty are not reported."
116
- ],
117
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry3",
118
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
119
  },
120
  {
121
  "item_id": "exp_3e426daf35ae",
122
  "mean": 63,
123
  "n": null,
124
  "uncertainty_status": "unknown",
125
- "uncertainty_scale": null,
126
- "uncertainty_basis": "No reliable per-condition difference uncertainty is reported.",
127
- "quality_status": "retained_with_flag",
128
- "warnings": [
129
- "Replication and difference uncertainty are not reported."
130
- ],
131
- "source_location": "Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry2",
132
- "aggregation": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
133
- }
134
- ],
135
- "pairs": [
136
- {
137
- "item_i": "exp_2dfe2f22dd32",
138
- "item_j": "exp_34e91dee44f6",
139
- "mean_difference": -1,
140
- "difference_se": null,
141
- "uncertainty_status": "unknown",
142
- "comparison_basis": "reported_value",
143
- "comparable": 1
144
- },
145
- {
146
- "item_i": "exp_2dfe2f22dd32",
147
- "item_j": "exp_0003b83bcce3",
148
- "mean_difference": 19,
149
- "difference_se": null,
150
- "uncertainty_status": "unknown",
151
- "comparison_basis": "reported_value",
152
- "comparable": 1
153
- },
154
- {
155
- "item_i": "exp_2dfe2f22dd32",
156
- "item_j": "exp_595fa6ce8e2c",
157
- "mean_difference": 13,
158
- "difference_se": null,
159
- "uncertainty_status": "unknown",
160
- "comparison_basis": "reported_value",
161
- "comparable": 1
162
- },
163
- {
164
- "item_i": "exp_2dfe2f22dd32",
165
- "item_j": "exp_1da97213829d",
166
- "mean_difference": 1,
167
- "difference_se": null,
168
- "uncertainty_status": "unknown",
169
- "comparison_basis": "reported_value",
170
- "comparable": 1
171
- },
172
- {
173
- "item_i": "exp_2dfe2f22dd32",
174
- "item_j": "exp_2da4d7cbbf59",
175
- "mean_difference": 0,
176
- "difference_se": null,
177
- "uncertainty_status": "unknown",
178
- "comparison_basis": "reported_value",
179
- "comparable": 0
180
- },
181
- {
182
- "item_i": "exp_2dfe2f22dd32",
183
- "item_j": "exp_2191e72320e5",
184
- "mean_difference": -2,
185
- "difference_se": null,
186
- "uncertainty_status": "unknown",
187
- "comparison_basis": "reported_value",
188
- "comparable": 1
189
- },
190
- {
191
- "item_i": "exp_2dfe2f22dd32",
192
- "item_j": "exp_1638cbaa31ee",
193
- "mean_difference": -3,
194
- "difference_se": null,
195
- "uncertainty_status": "unknown",
196
- "comparison_basis": "reported_value",
197
- "comparable": 1
198
- },
199
- {
200
- "item_i": "exp_2dfe2f22dd32",
201
- "item_j": "exp_3e426daf35ae",
202
- "mean_difference": -4,
203
- "difference_se": null,
204
- "uncertainty_status": "unknown",
205
- "comparison_basis": "reported_value",
206
- "comparable": 1
207
  },
208
  {
209
- "item_i": "exp_34e91dee44f6",
210
- "item_j": "exp_0003b83bcce3",
211
- "mean_difference": 20,
212
- "difference_se": null,
213
- "uncertainty_status": "unknown",
214
- "comparison_basis": "reported_value",
215
- "comparable": 1
216
- },
217
- {
218
- "item_i": "exp_34e91dee44f6",
219
- "item_j": "exp_595fa6ce8e2c",
220
- "mean_difference": 14,
221
- "difference_se": null,
222
- "uncertainty_status": "unknown",
223
- "comparison_basis": "reported_value",
224
- "comparable": 1
225
- },
226
- {
227
- "item_i": "exp_34e91dee44f6",
228
- "item_j": "exp_1da97213829d",
229
- "mean_difference": 2,
230
- "difference_se": null,
231
- "uncertainty_status": "unknown",
232
- "comparison_basis": "reported_value",
233
- "comparable": 1
234
- },
235
- {
236
- "item_i": "exp_34e91dee44f6",
237
- "item_j": "exp_2da4d7cbbf59",
238
- "mean_difference": 1,
239
- "difference_se": null,
240
- "uncertainty_status": "unknown",
241
- "comparison_basis": "reported_value",
242
- "comparable": 1
243
- },
244
- {
245
- "item_i": "exp_34e91dee44f6",
246
- "item_j": "exp_2191e72320e5",
247
- "mean_difference": -1,
248
- "difference_se": null,
249
- "uncertainty_status": "unknown",
250
- "comparison_basis": "reported_value",
251
- "comparable": 1
252
- },
253
- {
254
- "item_i": "exp_34e91dee44f6",
255
- "item_j": "exp_1638cbaa31ee",
256
- "mean_difference": -2,
257
- "difference_se": null,
258
- "uncertainty_status": "unknown",
259
- "comparison_basis": "reported_value",
260
- "comparable": 1
261
- },
262
- {
263
- "item_i": "exp_34e91dee44f6",
264
- "item_j": "exp_3e426daf35ae",
265
- "mean_difference": -3,
266
- "difference_se": null,
267
- "uncertainty_status": "unknown",
268
- "comparison_basis": "reported_value",
269
- "comparable": 1
270
- },
271
- {
272
- "item_i": "exp_0003b83bcce3",
273
- "item_j": "exp_595fa6ce8e2c",
274
- "mean_difference": -6,
275
- "difference_se": null,
276
- "uncertainty_status": "unknown",
277
- "comparison_basis": "reported_value",
278
- "comparable": 1
279
- },
280
- {
281
- "item_i": "exp_0003b83bcce3",
282
- "item_j": "exp_1da97213829d",
283
- "mean_difference": -18,
284
- "difference_se": null,
285
- "uncertainty_status": "unknown",
286
- "comparison_basis": "reported_value",
287
- "comparable": 1
288
- },
289
- {
290
- "item_i": "exp_0003b83bcce3",
291
- "item_j": "exp_2da4d7cbbf59",
292
- "mean_difference": -19,
293
- "difference_se": null,
294
- "uncertainty_status": "unknown",
295
- "comparison_basis": "reported_value",
296
- "comparable": 1
297
- },
298
- {
299
- "item_i": "exp_0003b83bcce3",
300
- "item_j": "exp_2191e72320e5",
301
- "mean_difference": -21,
302
- "difference_se": null,
303
- "uncertainty_status": "unknown",
304
- "comparison_basis": "reported_value",
305
- "comparable": 1
306
- },
307
- {
308
- "item_i": "exp_0003b83bcce3",
309
- "item_j": "exp_1638cbaa31ee",
310
- "mean_difference": -22,
311
- "difference_se": null,
312
- "uncertainty_status": "unknown",
313
- "comparison_basis": "reported_value",
314
- "comparable": 1
315
- },
316
- {
317
- "item_i": "exp_0003b83bcce3",
318
- "item_j": "exp_3e426daf35ae",
319
- "mean_difference": -23,
320
- "difference_se": null,
321
- "uncertainty_status": "unknown",
322
- "comparison_basis": "reported_value",
323
- "comparable": 1
324
- },
325
- {
326
- "item_i": "exp_595fa6ce8e2c",
327
- "item_j": "exp_1da97213829d",
328
- "mean_difference": -12,
329
- "difference_se": null,
330
- "uncertainty_status": "unknown",
331
- "comparison_basis": "reported_value",
332
- "comparable": 1
333
- },
334
- {
335
- "item_i": "exp_595fa6ce8e2c",
336
- "item_j": "exp_2da4d7cbbf59",
337
- "mean_difference": -13,
338
- "difference_se": null,
339
- "uncertainty_status": "unknown",
340
- "comparison_basis": "reported_value",
341
- "comparable": 1
342
- },
343
- {
344
- "item_i": "exp_595fa6ce8e2c",
345
- "item_j": "exp_2191e72320e5",
346
- "mean_difference": -15,
347
- "difference_se": null,
348
- "uncertainty_status": "unknown",
349
- "comparison_basis": "reported_value",
350
- "comparable": 1
351
- },
352
- {
353
- "item_i": "exp_595fa6ce8e2c",
354
- "item_j": "exp_1638cbaa31ee",
355
- "mean_difference": -16,
356
- "difference_se": null,
357
- "uncertainty_status": "unknown",
358
- "comparison_basis": "reported_value",
359
- "comparable": 1
360
- },
361
- {
362
- "item_i": "exp_595fa6ce8e2c",
363
- "item_j": "exp_3e426daf35ae",
364
- "mean_difference": -17,
365
- "difference_se": null,
366
- "uncertainty_status": "unknown",
367
- "comparison_basis": "reported_value",
368
- "comparable": 1
369
- },
370
- {
371
- "item_i": "exp_1da97213829d",
372
- "item_j": "exp_2da4d7cbbf59",
373
- "mean_difference": -1,
374
- "difference_se": null,
375
- "uncertainty_status": "unknown",
376
- "comparison_basis": "reported_value",
377
- "comparable": 1
378
- },
379
- {
380
- "item_i": "exp_1da97213829d",
381
- "item_j": "exp_2191e72320e5",
382
- "mean_difference": -3,
383
- "difference_se": null,
384
- "uncertainty_status": "unknown",
385
- "comparison_basis": "reported_value",
386
- "comparable": 1
387
- },
388
- {
389
- "item_i": "exp_1da97213829d",
390
- "item_j": "exp_1638cbaa31ee",
391
- "mean_difference": -4,
392
- "difference_se": null,
393
- "uncertainty_status": "unknown",
394
- "comparison_basis": "reported_value",
395
- "comparable": 1
396
- },
397
- {
398
- "item_i": "exp_1da97213829d",
399
- "item_j": "exp_3e426daf35ae",
400
- "mean_difference": -5,
401
- "difference_se": null,
402
- "uncertainty_status": "unknown",
403
- "comparison_basis": "reported_value",
404
- "comparable": 1
405
- },
406
- {
407
- "item_i": "exp_2da4d7cbbf59",
408
- "item_j": "exp_2191e72320e5",
409
- "mean_difference": -2,
410
- "difference_se": null,
411
- "uncertainty_status": "unknown",
412
- "comparison_basis": "reported_value",
413
- "comparable": 1
414
- },
415
- {
416
- "item_i": "exp_2da4d7cbbf59",
417
- "item_j": "exp_1638cbaa31ee",
418
- "mean_difference": -3,
419
- "difference_se": null,
420
- "uncertainty_status": "unknown",
421
- "comparison_basis": "reported_value",
422
- "comparable": 1
423
- },
424
- {
425
- "item_i": "exp_2da4d7cbbf59",
426
- "item_j": "exp_3e426daf35ae",
427
- "mean_difference": -4,
428
- "difference_se": null,
429
- "uncertainty_status": "unknown",
430
- "comparison_basis": "reported_value",
431
- "comparable": 1
432
- },
433
- {
434
- "item_i": "exp_2191e72320e5",
435
- "item_j": "exp_1638cbaa31ee",
436
- "mean_difference": -1,
437
- "difference_se": null,
438
- "uncertainty_status": "unknown",
439
- "comparison_basis": "reported_value",
440
- "comparable": 1
441
- },
442
- {
443
- "item_i": "exp_2191e72320e5",
444
- "item_j": "exp_3e426daf35ae",
445
- "mean_difference": -2,
446
- "difference_se": null,
447
- "uncertainty_status": "unknown",
448
- "comparison_basis": "reported_value",
449
- "comparable": 1
450
- },
451
- {
452
- "item_i": "exp_1638cbaa31ee",
453
- "item_j": "exp_3e426daf35ae",
454
- "mean_difference": -1,
455
- "difference_se": null,
456
  "uncertainty_status": "unknown",
457
- "comparison_basis": "reported_value",
458
- "comparable": 1
459
  }
460
- ],
461
- "comparable_pairs": 35,
462
- "quality_status": "passed"
463
  }
464
  ]
465
  }
 
1
  {
2
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "reaction_settings",
7
  "items": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  {
9
  "item_id": "exp_0003b83bcce3",
10
  "mean": 40,
11
  "n": null,
12
  "uncertainty_status": "unknown",
13
+ "uncertainty_scale": null
 
 
 
 
 
 
 
14
  },
15
  {
16
+ "item_id": "exp_1638cbaa31ee",
17
+ "mean": 62,
18
  "n": null,
19
  "uncertainty_status": "unknown",
20
+ "uncertainty_scale": null
 
 
 
 
 
 
 
21
  },
22
  {
23
  "item_id": "exp_1da97213829d",
24
  "mean": 58,
25
  "n": null,
26
  "uncertainty_status": "unknown",
27
+ "uncertainty_scale": null
 
 
 
 
 
 
 
28
  },
29
  {
30
+ "item_id": "exp_2191e72320e5",
31
+ "mean": 61,
32
  "n": null,
33
  "uncertainty_status": "unknown",
34
+ "uncertainty_scale": null
 
 
 
 
 
 
 
35
  },
36
  {
37
+ "item_id": "exp_2dfe2f22dd32",
38
+ "mean": 59,
39
  "n": null,
40
  "uncertainty_status": "unknown",
41
+ "uncertainty_scale": null
 
 
 
 
 
 
 
42
  },
43
  {
44
+ "item_id": "exp_34e91dee44f6",
45
+ "mean": 60,
46
  "n": null,
47
  "uncertainty_status": "unknown",
48
+ "uncertainty_scale": null
 
 
 
 
 
 
 
49
  },
50
  {
51
  "item_id": "exp_3e426daf35ae",
52
  "mean": 63,
53
  "n": null,
54
  "uncertainty_status": "unknown",
55
+ "uncertainty_scale": null
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
56
  },
57
  {
58
+ "item_id": "exp_595fa6ce8e2c",
59
+ "mean": 46,
60
+ "n": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
  "uncertainty_status": "unknown",
62
+ "uncertainty_scale": null
 
63
  }
64
+ ]
 
 
65
  }
66
  ]
67
  }
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/contract.json CHANGED
@@ -1,33 +1,27 @@
1
  {
2
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "reaction_settings",
7
  "label": "Reaction settings",
8
  "item_ids": [
9
- "exp_2dfe2f22dd32",
10
- "exp_34e91dee44f6",
11
  "exp_0003b83bcce3",
12
- "exp_595fa6ce8e2c",
13
  "exp_1da97213829d",
14
- "exp_2da4d7cbbf59",
15
  "exp_2191e72320e5",
16
- "exp_1638cbaa31ee",
17
- "exp_3e426daf35ae"
 
 
18
  ],
19
  "direction": "ascending",
20
  "group_weight": 1.0,
21
- "k": 2,
22
- "epsilon": 0,
23
- "uncertainty_model": "optional_reported_uncertainty",
24
- "difference_standard_error": "sem_sum_upper_bound",
25
  "endpoint": {
26
  "name": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard.",
27
  "unit": "%",
28
  "definition_en": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
29
- },
30
- "uncertainty_definition": "Independent and technical replicate counts and reliable difference uncertainties are not reported for this panel. All errors remain unknown; the reported-value branch is used."
31
  }
32
  ],
33
  "scoring_method": "spearman",
@@ -39,18 +33,9 @@
39
  "spearman_tie_method": "average",
40
  "spearman_undefined_policy": "task_error",
41
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
42
- "secondary_scoring_method": "equal_pair_inversion_random95",
43
- "ranking_reward_range": [
44
- 0,
45
- 1
46
- ],
47
- "random_quantile": 0.95,
48
- "unknown_uncertainty_policy": "use_reported_value",
49
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
50
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
51
  "format_errors": {
52
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
53
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
54
  },
55
  "max_submission_bytes": 1048576,
56
  "minimum_conditions_per_group": 2,
@@ -68,6 +53,13 @@
68
  "sources_path": "/logs/artifacts/sources.json",
69
  "analysis_path": "/logs/artifacts/analysis/",
70
  "forbidden_result_scope": "Same-type forbidden results are nickel-mediated asymmetric difluoromethylations of unactivated alkyl halides with zinc difluoromethyl reagents, including yields, conversion, enantiomeric excess, additive/ligand/nickel/oxidant/solvent comparisons and outcome-fitted predictors.",
71
- "pair_boundary_arithmetic": "exact_decimal",
72
- "pair_boundary_definition": "Interpret each stored numeric input by its decimal text and evaluate the strict pair inequality exactly. Equality is not a decidable pair. For independent SEM quadrature, compare positive squared sides exactly; no floating-point tolerance is added to k or epsilon."
 
 
 
 
 
 
 
73
  }
 
1
  {
2
  "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "reaction_settings",
7
  "label": "Reaction settings",
8
  "item_ids": [
 
 
9
  "exp_0003b83bcce3",
10
+ "exp_1638cbaa31ee",
11
  "exp_1da97213829d",
 
12
  "exp_2191e72320e5",
13
+ "exp_2dfe2f22dd32",
14
+ "exp_34e91dee44f6",
15
+ "exp_3e426daf35ae",
16
+ "exp_595fa6ce8e2c"
17
  ],
18
  "direction": "ascending",
19
  "group_weight": 1.0,
 
 
 
 
20
  "endpoint": {
21
  "name": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard.",
22
  "unit": "%",
23
  "definition_en": "19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
24
+ }
 
25
  }
26
  ],
27
  "scoring_method": "spearman",
 
33
  "spearman_tie_method": "average",
34
  "spearman_undefined_policy": "task_error",
35
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
36
  "format_errors": {
37
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
38
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
39
  },
40
  "max_submission_bytes": 1048576,
41
  "minimum_conditions_per_group": 2,
 
53
  "sources_path": "/logs/artifacts/sources.json",
54
  "analysis_path": "/logs/artifacts/analysis/",
55
  "forbidden_result_scope": "Same-type forbidden results are nickel-mediated asymmetric difluoromethylations of unactivated alkyl halides with zinc difluoromethyl reagents, including yields, conversion, enantiomeric excess, additive/ligand/nickel/oxidant/solvent comparisons and outcome-fitted predictors.",
56
+ "supporting_artifact_review": {
57
+ "policy_stop_marker": "SPU_POLICY_STOP",
58
+ "required_files": [
59
+ "report.md",
60
+ "sources.json"
61
+ ],
62
+ "required_directory": "analysis",
63
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
64
+ }
65
  }
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/grade.py CHANGED
@@ -1,14 +1,44 @@
1
- """Pipeline v18 Spearman core and preserved equal-pair auxiliary grader."""
2
  import argparse
3
- import copy
4
  import hashlib
5
  import json
6
  import math
 
 
7
  from pathlib import Path
8
 
9
- import original_grade as old
10
 
11
- TaskError = old.TaskError
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
 
13
 
14
  def bounded(value):
@@ -51,98 +81,174 @@ def spearman(values, predicted_positions):
51
  return bounded(math.fsum(r * p for r, p in zip(x, y)) / denominator)
52
 
53
 
54
- def promote_contract(contract):
55
- value = copy.deepcopy(contract)
56
- value.update(scoring_method='spearman', primary_score='task_spearman',
57
- spearman_range=[-1, 1], spearman_tie_method='average',
58
- spearman_undefined_policy='task_error',
59
- secondary_scoring_method='equal_pair_inversion_random95',
60
- ranking_reward_range=[0, 1])
61
- return value
62
-
63
-
64
- def auxiliary_view(contract):
65
- expected = {'scoring_method': 'spearman', 'primary_score': 'task_spearman',
66
- 'spearman_range': [-1, 1], 'spearman_tie_method': 'average',
67
- 'spearman_undefined_policy': 'task_error',
68
- 'secondary_scoring_method': 'equal_pair_inversion_random95',
69
- 'ranking_reward_range': [0, 1]}
70
- if any(contract.get(k) != v for k, v in expected.items()):
71
- raise TaskError('Invalid dual-score contract')
72
- value = copy.deepcopy(contract)
73
- value['scoring_method'] = value.pop('secondary_scoring_method')
74
- for key in ('primary_score', 'spearman_range', 'spearman_tie_method',
75
- 'spearman_undefined_policy', 'ranking_reward_range'):
76
- value.pop(key)
77
- return value
78
-
79
-
80
- def scoring_bindings(contract, answers):
81
- return {'contract_sha256': old.canonical_sha(contract),
82
- 'answers_sha256': old.canonical_sha(answers),
83
- 'grader_sha256': hashlib.sha256(Path(__file__).read_bytes()).hexdigest(),
84
- 'auxiliary_grader_sha256': hashlib.sha256(Path(old.__file__).read_bytes()).hexdigest()}
85
-
86
-
87
- def validate(contract, answers, baseline):
88
- auxiliary = auxiliary_view(contract)
89
- old.validate_task(auxiliary, answers)
90
- if baseline.get('bindings') != scoring_bindings(contract, answers):
91
- raise TaskError('Dual-score baseline hash mismatch')
92
- old_baseline = {'q': baseline.get('q'), 'bindings': old.scoring_bindings(auxiliary, answers)}
93
- old.validate_baseline(auxiliary, answers, old_baseline)
94
- hidden = {g['group_id']: {i['item_id']: i['mean'] for i in g['items']} for g in answers['groups']}
95
- for g in contract['groups']:
96
- spearman([hidden[g['group_id']][i] for i in g['item_ids']], list(range(len(g['item_ids']))))
97
- return auxiliary, old_baseline, hidden
98
-
99
-
100
- def add_core(contract, report, hidden):
101
- report['scoring_method'] = 'spearman'
102
- report['primary_score'] = 'task_spearman'
103
- report['secondary_scoring_method'] = 'equal_pair_inversion_random95'
104
- report['official_reward'] = None
105
- report['official_ranking_reward'] = None
106
- if not report['groups']:
107
- report['task_spearman'] = -1.0
108
- report['core_penalty_reason'] = 'whole_task_format_error'
109
- report['groups'] = [{'group_id': g['group_id'], 'group_weight': g['group_weight'],
110
- 'group_spearman_g': None, 'spearman_for_aggregation': -1.0,
111
- 'group_pair_accuracy': 0.0, 'status': 'whole_task_format_error'}
112
- for g in contract['groups']]
113
- return report
114
- for g in report['groups']:
115
- if g['status'] != 'valid':
116
- rho = None
117
- else:
118
- item_ids = next(c['item_ids'] for c in contract['groups'] if c['group_id'] == g['group_id'])
119
- positions = {item_id: i for i, item_id in enumerate(g['order'])}
120
- rho = spearman([hidden[g['group_id']][i] for i in item_ids], [positions[i] for i in item_ids])
121
- g['group_spearman_g'] = rho
122
- g['spearman_for_aggregation'] = -1.0 if rho is None else rho
123
- report['task_spearman'] = bounded(math.fsum(g['group_weight'] * g['spearman_for_aggregation']
124
- for g in report['groups']))
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
125
  return report
126
 
127
 
128
- def grade_submission(contract, answers, submission, baseline):
129
- auxiliary, old_baseline, hidden = validate(contract, answers, baseline)
130
- report = old.grade_submission(auxiliary, answers, submission, old_baseline)
131
- report['random_q95'] = baseline['q']
132
- return add_core(contract, report, hidden)
133
 
134
 
135
- def grade_file(contract, answers, path, baseline):
136
- auxiliary, old_baseline, hidden = validate(contract, answers, baseline)
137
- return add_core(contract, old.grade_file(auxiliary, answers, path, old_baseline), hidden)
138
 
 
 
 
 
 
 
 
 
 
 
 
 
139
 
140
- def official_scores(report, compliance, execution, memory=False):
141
- if compliance == 'cheating':
142
- return -1.0, 0.0
143
- if compliance != 'compliant' or execution != 'completed':
144
- return None, None
145
- return report['task_spearman'], report['ranking_reward']
 
 
 
 
 
 
 
146
 
147
 
148
  def main():
@@ -156,18 +262,22 @@ def main():
156
  for name in ('reward.txt', 'reward.json', 'report.json', 'evaluator_error.json'):
157
  (args.output / name).unlink(missing_ok=True)
158
  try:
159
- contract, answers, baseline = [old.load_json((args.root / f'{name}.json').read_bytes())
160
- for name in ('contract', 'answers', 'random_baseline')]
161
- report = grade_file(contract, answers, args.submission, baseline)
162
- except (TaskError, old.SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
163
- old.dump(args.output / 'evaluator_error.json', {'status': 'task_error', 'error': str(error)})
 
 
 
 
164
  raise SystemExit(2) from error
165
- old.dump(args.output / 'report.json', report)
 
166
  if args.interface == 'txt':
167
  (args.output / 'reward.txt').write_text(str(report['task_spearman']) + '\n')
168
  else:
169
- old.dump(args.output / 'reward.json', {'reward': report['task_spearman'],
170
- 'ranking_reward': report['ranking_reward']})
171
 
172
 
173
  if __name__ == '__main__':
 
1
+ """Score complete experimental rankings by weighted Spearman correlation."""
2
  import argparse
3
+ import errno
4
  import hashlib
5
  import json
6
  import math
7
+ import os
8
+ import stat
9
  from pathlib import Path
10
 
 
11
 
12
+ class TaskError(ValueError):
13
+ pass
14
+
15
+
16
+ class SubmissionError(ValueError):
17
+ pass
18
+
19
+
20
+ def unique_object(pairs):
21
+ result = {}
22
+ for key, value in pairs:
23
+ if key in result:
24
+ raise SubmissionError("duplicate_json_key")
25
+ result[key] = value
26
+ return result
27
+
28
+
29
+ def load_json(raw):
30
+ def reject_constant(value):
31
+ raise SubmissionError("nonfinite_json_constant")
32
+ return json.loads(raw.decode("utf-8"), object_pairs_hook=unique_object,
33
+ parse_constant=reject_constant)
34
+
35
+
36
+ def finite(value, name, minimum=None):
37
+ if isinstance(value, bool) or not isinstance(value, (int, float)):
38
+ raise TaskError(name + " is not numeric")
39
+ if not math.isfinite(value) or (minimum is not None and value < minimum):
40
+ raise TaskError(name + " is out of range")
41
+ return float(value)
42
 
43
 
44
  def bounded(value):
 
81
  return bounded(math.fsum(r * p for r, p in zip(x, y)) / denominator)
82
 
83
 
84
+ def read_regular(path, limit):
85
+ try:
86
+ fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
87
+ try:
88
+ info = os.fstat(fd)
89
+ if not stat.S_ISREG(info.st_mode):
90
+ raise SubmissionError("not_regular_file")
91
+ if info.st_size > limit:
92
+ raise SubmissionError("submission_too_large")
93
+ with os.fdopen(fd, "rb", closefd=False) as stream:
94
+ raw = stream.read(limit + 1)
95
+ if len(raw) > limit:
96
+ raise SubmissionError("submission_too_large")
97
+ return raw
98
+ finally:
99
+ os.close(fd)
100
+ except OSError as error:
101
+ if error.errno in (errno.ENOENT, errno.ELOOP, errno.EISDIR, errno.ENOTDIR,
102
+ errno.ENXIO, errno.EACCES, errno.EPERM):
103
+ raise SubmissionError("missing_or_unreadable_submission") from error
104
+ raise
105
+
106
+
107
+ def validate_task(contract, answers):
108
+ expected = {
109
+ "scoring_method": "spearman",
110
+ "primary_score": "task_spearman",
111
+ "spearman_range": [-1, 1],
112
+ "spearman_tie_method": "average",
113
+ "spearman_undefined_policy": "task_error",
114
+ }
115
+ if any(contract.get(key) != value for key, value in expected.items()):
116
+ raise TaskError("Unsupported scoring contract")
117
+ for key in ("task_id", "task_version"):
118
+ if not isinstance(contract.get(key), str) or answers.get(key) != contract[key]:
119
+ raise TaskError("Invalid task identity")
120
+ limit = contract.get("max_submission_bytes")
121
+ if type(limit) is not int or not 0 < limit <= 1048576:
122
+ raise TaskError("Invalid submission size limit")
123
+ minimum = contract.get("minimum_conditions_per_group", 2)
124
+ if type(minimum) is not int or minimum < 2:
125
+ raise TaskError("Invalid minimum group size")
126
+ groups = contract["groups"]
127
+ references = {g["group_id"]: g for g in answers["groups"]}
128
+ if not groups or len(groups) != len(references) or len(references) != len(answers["groups"]):
129
+ raise TaskError("Invalid reference groups")
130
+
131
+ measurements = {}
132
+ all_ids = set()
133
+ for group in groups:
134
+ gid = group["group_id"]
135
+ ids = group["item_ids"]
136
+ if not isinstance(gid, str) or not gid or gid in measurements or gid not in references:
137
+ raise TaskError("Invalid group ID")
138
+ if (group["direction"] != "ascending" or not isinstance(ids, list) or len(ids) < minimum
139
+ or any(not isinstance(i, str) or not i for i in ids)
140
+ or len(set(ids)) != len(ids) or all_ids.intersection(ids)):
141
+ raise TaskError("Invalid group definition")
142
+ all_ids.update(ids)
143
+ if finite(group["group_weight"], "group_weight", 0) == 0:
144
+ raise TaskError("Zero group weight")
145
+ items = references[gid]["items"]
146
+ values = {item["item_id"]: item["mean"] for item in items}
147
+ if len(items) != len(ids) or set(values) != set(ids):
148
+ raise TaskError("Reference items differ from the contract")
149
+ measurements[gid] = [values[i] for i in ids]
150
+ spearman(measurements[gid], list(range(len(ids))))
151
+ if not math.isclose(math.fsum(g["group_weight"] for g in groups), 1.0, rel_tol=0, abs_tol=1e-12):
152
+ raise TaskError("Group weights must sum to one")
153
+ return measurements
154
+
155
+
156
+ def group_entries(submission, contract):
157
+ if not isinstance(submission, dict):
158
+ raise SubmissionError("invalid_top_level_structure")
159
+ if any(submission.get(key) != contract[key] for key in ("task_id", "task_version")):
160
+ raise SubmissionError("task_identity_mismatch")
161
+ if not isinstance(submission.get("rankings"), list):
162
+ raise SubmissionError("rankings_not_array")
163
+ entries = {group["group_id"]: [] for group in contract["groups"]}
164
+ for entry in submission["rankings"]:
165
+ if not isinstance(entry, dict) or not isinstance(entry.get("group_id"), str):
166
+ raise SubmissionError("unassignable_group_entry")
167
+ if entry["group_id"] not in entries:
168
+ raise SubmissionError("unknown_group_id")
169
+ entries[entry["group_id"]].append(entry)
170
+ return entries
171
+
172
+
173
+ def score_group(group, entries, values):
174
+ gid = group["group_id"]
175
+ ids = group["item_ids"]
176
+ result = {"group_id": gid, "group_weight": group["group_weight"],
177
+ "group_spearman_g": None, "spearman_for_aggregation": -1.0}
178
+ if entries is None:
179
+ result["status"] = "whole_task_format_error"
180
+ return result
181
+ matches = entries[gid]
182
+ if len(matches) != 1:
183
+ result["status"] = "duplicate_group" if matches else "missing_group"
184
+ return result
185
+ order = matches[0].get("order")
186
+ if (not isinstance(order, list) or len(order) != len(ids)
187
+ or any(not isinstance(i, str) for i in order) or set(order) != set(ids)):
188
+ result["status"] = "invalid_permutation"
189
+ return result
190
+ positions = {item: p for p, item in enumerate(order)}
191
+ rho = spearman(values, [positions[i] for i in ids])
192
+ result.update(status="valid", order=order, group_spearman_g=rho, spearman_for_aggregation=rho)
193
+ return result
194
+
195
+
196
+ def grade_file(contract, answers, path):
197
+ measurements = validate_task(contract, answers)
198
+ report = {"task_id": contract["task_id"], "task_version": contract["task_version"],
199
+ "scoring_method": "spearman", "primary_score": "task_spearman",
200
+ "evaluation_status": "completed"}
201
+ entries = None
202
+ try:
203
+ raw = read_regular(path, contract["max_submission_bytes"])
204
+ submission = load_json(raw)
205
+ report["submission_sha256"] = hashlib.sha256(raw).hexdigest()
206
+ entries = group_entries(submission, contract)
207
+ except (SubmissionError, UnicodeError, json.JSONDecodeError, RecursionError) as error:
208
+ report["submission_status"] = str(error) or type(error).__name__
209
+ report["core_penalty_reason"] = "whole_task_format_error"
210
+ groups = [score_group(g, entries, measurements[g["group_id"]]) for g in contract["groups"]]
211
+ report["groups"] = groups
212
+ if entries is None:
213
+ report["task_spearman"] = -1.0
214
+ else:
215
+ report["submission_status"] = "valid" if all(g["status"] == "valid" for g in groups) else "group_errors"
216
+ report["task_spearman"] = bounded(math.fsum(g["group_weight"] * g["spearman_for_aggregation"]
217
+ for g in groups))
218
  return report
219
 
220
 
221
+ def dump(path, value):
222
+ path.write_text(json.dumps(value, indent=2, allow_nan=False) + "\n", encoding="utf-8")
 
 
 
223
 
224
 
 
 
 
225
 
226
+ def supporting_artifact_status(submission):
227
+ root = submission.parent
228
+ report_path = root / 'report.md'
229
+ paths = {'report_file': report_path, 'sources_file': root / 'sources.json',
230
+ 'analysis_directory': root / 'analysis'}
231
+ present = {}
232
+ for name, path in paths.items():
233
+ try:
234
+ mode = path.lstat().st_mode
235
+ present[name] = stat.S_ISDIR(mode) if name == 'analysis_directory' else stat.S_ISREG(mode)
236
+ except OSError:
237
+ present[name] = False
238
 
239
+ declared_stop = False
240
+ if present['report_file']:
241
+ try:
242
+ fd = os.open(report_path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
243
+ with os.fdopen(fd, 'rb') as stream:
244
+ if stat.S_ISREG(os.fstat(stream.fileno()).st_mode):
245
+ declared_stop = stream.readline(256).rstrip(b'\r\n') == b'SPU_POLICY_STOP'
246
+ except OSError:
247
+ pass
248
+ status = 'artifacts_present' if all(present.values()) else 'artifacts_incomplete'
249
+ if declared_stop:
250
+ status = 'declared_stop_with_ranking' if os.path.lexists(submission) else 'declared_stop'
251
+ return {'status': status, **present, 'policy_stop_declared': declared_stop}
252
 
253
 
254
  def main():
 
262
  for name in ('reward.txt', 'reward.json', 'report.json', 'evaluator_error.json'):
263
  (args.output / name).unlink(missing_ok=True)
264
  try:
265
+ bindings = load_json((args.root / 'scoring_hashes.json').read_bytes())
266
+ for name in ('contract.json', 'answers.json', 'grade.py'):
267
+ if hashlib.sha256((args.root / name).read_bytes()).hexdigest() != bindings[name]:
268
+ raise TaskError('Scoring file hash mismatch: ' + name)
269
+ contract = load_json((args.root / 'contract.json').read_bytes())
270
+ answers = load_json((args.root / 'answers.json').read_bytes())
271
+ report = grade_file(contract, answers, args.submission)
272
+ except (TaskError, SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
273
+ dump(args.output / 'evaluator_error.json', {'status': 'task_error', 'error': str(error)})
274
  raise SystemExit(2) from error
275
+ report['supporting_artifact_review'] = supporting_artifact_status(args.submission)
276
+ dump(args.output / 'report.json', report)
277
  if args.interface == 'txt':
278
  (args.output / 'reward.txt').write_text(str(report['task_spearman']) + '\n')
279
  else:
280
+ dump(args.output / 'reward.json', {'reward': report['task_spearman']})
 
281
 
282
 
283
  if __name__ == '__main__':
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/measurements.csv CHANGED
@@ -1,10 +1,9 @@
1
  group_id,item_id,mean,n,uncertainty_status,uncertainty_scale,uncertainty_basis,quality_status,warnings,source_location,aggregation
2
- reaction_settings,exp_2dfe2f22dd32,59,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry6","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
3
- reaction_settings,exp_34e91dee44f6,60,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry1","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
4
  reaction_settings,exp_0003b83bcce3,40,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry8","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
5
- reaction_settings,exp_595fa6ce8e2c,46,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry4","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
6
  reaction_settings,exp_1da97213829d,58,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry9","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
7
- reaction_settings,exp_2da4d7cbbf59,59,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry5","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
8
  reaction_settings,exp_2191e72320e5,61,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry7","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
9
- reaction_settings,exp_1638cbaa31ee,62,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry3","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
 
10
  reaction_settings,exp_3e426daf35ae,63,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry2","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
 
 
1
  group_id,item_id,mean,n,uncertainty_status,uncertainty_scale,uncertainty_basis,quality_status,warnings,source_location,aggregation
 
 
2
  reaction_settings,exp_0003b83bcce3,40,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry8","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
3
+ reaction_settings,exp_1638cbaa31ee,62,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry3","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
4
  reaction_settings,exp_1da97213829d,58,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry9","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
 
5
  reaction_settings,exp_2191e72320e5,61,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry7","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
6
+ reaction_settings,exp_2dfe2f22dd32,59,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry6","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
7
+ reaction_settings,exp_34e91dee44f6,60,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry1","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
8
  reaction_settings,exp_3e426daf35ae,63,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry2","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
9
+ reaction_settings,exp_595fa6ce8e2c,46,,unknown,,No reliable per-condition difference uncertainty is reported.,retained_with_flag,"[""Replication and difference uncertainty are not reported.""]","Original-SI.pdf physical p46(printed45), Supplementary Table15 complete9-entry additive panel,entry4","19F NMR percentage yield of the specified difluoromethylated amide relative to 0.05 mmol alkyl iodide, with ethyl bromodifluoroacetate internal standard."
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/original_grade.py DELETED
@@ -1,451 +0,0 @@
1
- """Pipeline v16: equal pairs within groups; one task-level random-95 threshold."""
2
-
3
- import argparse
4
- import errno
5
- import hashlib
6
- import itertools
7
- import json
8
- import math
9
- import os
10
- from pathlib import Path
11
- import stat
12
-
13
-
14
- class TaskError(ValueError):
15
- pass
16
-
17
-
18
- class SubmissionError(ValueError):
19
- pass
20
-
21
-
22
- def canonical_sha(value):
23
- return hashlib.sha256(json.dumps(value, sort_keys=True, separators=(",", ":"),
24
- ensure_ascii=False, allow_nan=False).encode()).hexdigest()
25
-
26
-
27
- def scoring_bindings(contract, answers):
28
- return {
29
- "contract_sha256": canonical_sha(contract),
30
- "answers_sha256": canonical_sha(answers),
31
- "relations_sha256": canonical_sha([{"group_id": g["group_id"], "pairs": g["pairs"]}
32
- for g in answers["groups"]]),
33
- "weights_sha256": canonical_sha([{k: g[k] for k in ("group_id", "group_weight")}
34
- for g in contract["groups"]]),
35
- "grader_sha256": hashlib.sha256(Path(__file__).read_bytes()).hexdigest(),
36
- }
37
-
38
-
39
- def validate_baseline(contract, answers, baseline):
40
- if not isinstance(baseline, dict):
41
- raise TaskError("random baseline must be an object")
42
- q = finite(baseline.get("q"), "random_q95", 0.5)
43
- if q >= 1:
44
- raise TaskError("random_q95 must be below 1")
45
- if contract.get("random_quantile") != 0.95:
46
- raise TaskError("invalid random quantile")
47
- if baseline.get("bindings") != scoring_bindings(contract, answers):
48
- raise TaskError("random baseline hash mismatch")
49
- return q
50
-
51
-
52
- def reward_from_accuracy(accuracy, q):
53
- accuracy = finite(accuracy, "accuracy", 0)
54
- q = finite(q, "random_q95", 0.5)
55
- if accuracy > 1 or q >= 1:
56
- raise TaskError("invalid accuracy or random threshold")
57
- if accuracy == 1:
58
- return 1.0
59
- if accuracy <= q + 1e-12:
60
- return 0.0
61
- return min(1.0, max(0.0, (accuracy - q) / (1 - q)))
62
-
63
-
64
- def aggregate_accuracy(details):
65
- # Preserve exact endpoints despite representational roundoff in weights.
66
- if all(g["group_pair_accuracy"] == 1 for g in details):
67
- return 1.0
68
- return math.fsum(g["group_pair_accuracy"] * g["group_weight"] for g in details)
69
-
70
-
71
- def unique_object(pairs):
72
- result = {}
73
- for key, value in pairs:
74
- if key in result:
75
- raise SubmissionError("duplicate_json_key")
76
- result[key] = value
77
- return result
78
-
79
-
80
- def load_json(raw):
81
- def reject_constant(value):
82
- raise SubmissionError("nonfinite_json_constant")
83
- return json.loads(raw.decode("utf-8"), object_pairs_hook=unique_object,
84
- parse_constant=reject_constant)
85
-
86
-
87
- def finite(value, name, minimum=None):
88
- if isinstance(value, bool) or not isinstance(value, (int, float)):
89
- raise TaskError(name + " is not numeric")
90
- if not math.isfinite(value) or (minimum is not None and value < minimum):
91
- raise TaskError(name + " is out of range")
92
- return float(value)
93
-
94
-
95
- def difference_se(sem_i, sem_j, method):
96
- if method == "sem_sum_upper_bound":
97
- return sem_i + sem_j
98
- if method == "independent_sem_quadrature":
99
- return math.hypot(sem_i, sem_j)
100
- raise TaskError("unsupported difference standard error method")
101
-
102
-
103
- def pair_comparable(mean_i, mean_j, sem_i, sem_j, method, k, epsilon):
104
- """Compare the exact decimal values represented by validated JSON numbers.
105
-
106
- The positive-side squared test avoids rounding a quadrature square root.
107
- Float differences and scales remain diagnostics, not boundary decisions.
108
- """
109
- from fractions import Fraction
110
-
111
- remaining = abs(Fraction(str(mean_i)) - Fraction(str(mean_j))) - Fraction(str(epsilon))
112
- if remaining <= 0:
113
- return False
114
- if sem_i is None or sem_j is None:
115
- return True
116
- multiplier = Fraction(str(k))
117
- left_scale, right_scale = Fraction(str(sem_i)), Fraction(str(sem_j))
118
- if method == "sem_sum_upper_bound":
119
- return remaining > multiplier * (left_scale + right_scale)
120
- if method == "independent_sem_quadrature":
121
- return remaining * remaining > multiplier * multiplier * (left_scale * left_scale + right_scale * right_scale)
122
- raise TaskError("unsupported difference standard error method")
123
-
124
-
125
- def validate_task(contract, answers):
126
- if not isinstance(contract, dict) or not isinstance(answers, dict):
127
- raise TaskError("contract and answers must be objects")
128
- if contract.get("pair_boundary_arithmetic") != "exact_decimal":
129
- raise TaskError("exact decimal grader requires explicit contract arithmetic")
130
- if contract.get("scoring_method") != "equal_pair_inversion_random95":
131
- raise TaskError("unsupported scoring method")
132
- if contract.get("unknown_uncertainty_policy") != "use_reported_value":
133
- raise TaskError("unsupported unknown-uncertainty policy")
134
- for key in ("task_id", "task_version"):
135
- if not isinstance(contract.get(key), str) or answers.get(key) != contract[key]:
136
- raise TaskError("invalid task identity")
137
- limit = contract.get("max_submission_bytes")
138
- if type(limit) is not int or not 0 < limit <= 1048576:
139
- raise TaskError("invalid submission size limit")
140
- minimum_conditions = contract.get("minimum_conditions_per_group", 2)
141
- if type(minimum_conditions) is not int or minimum_conditions < 2:
142
- raise TaskError("invalid minimum conditions")
143
- groups = contract.get("groups")
144
- hidden_groups = answers.get("groups")
145
- if not isinstance(groups, list) or not groups or not isinstance(hidden_groups, list):
146
- raise TaskError("missing groups")
147
- if len(groups) != len(hidden_groups):
148
- raise TaskError("hidden group count mismatch")
149
- hidden_map = {}
150
- for hidden in hidden_groups:
151
- if not isinstance(hidden, dict) or not isinstance(hidden.get("group_id"), str):
152
- raise TaskError("invalid hidden group")
153
- if hidden["group_id"] in hidden_map:
154
- raise TaskError("duplicate hidden group")
155
- hidden_map[hidden["group_id"]] = hidden
156
- validated, group_ids, all_ids = [], set(), set()
157
- for group in groups:
158
- if not isinstance(group, dict):
159
- raise TaskError("invalid contract group")
160
- gid = group.get("group_id")
161
- if not isinstance(gid, str) or not gid or gid in group_ids:
162
- raise TaskError("invalid group id")
163
- group_ids.add(gid)
164
- ids = group.get("item_ids")
165
- if (not isinstance(ids, list) or len(ids) < minimum_conditions
166
- or any(not isinstance(item, str) or not item for item in ids)
167
- or len(ids) != len(set(ids)) or all_ids.intersection(ids)):
168
- raise TaskError("invalid item ids")
169
- all_ids.update(ids)
170
- if group.get("direction") != "ascending":
171
- raise TaskError("only ascending order is supported")
172
- group_weight = finite(group.get("group_weight"), "group_weight", 0)
173
- if group_weight == 0:
174
- raise TaskError("zero group weight")
175
- k = finite(group.get("k"), "k", 0)
176
- epsilon = finite(group.get("epsilon"), "epsilon", 0)
177
- method = group.get("difference_standard_error")
178
- if method not in ("sem_sum_upper_bound", "independent_sem_quadrature", "unknown"):
179
- raise TaskError("unsupported difference standard error method")
180
- if group.get("uncertainty_model") not in ("optional_reported_uncertainty",
181
- "reported_sd_with_source_bounds", "independent_replicate_mean", "replicate_sd_upper_envelope", "unknown_replicate_hierarchy"):
182
- raise TaskError("unsupported marginal uncertainty model")
183
- hidden = hidden_map.get(gid)
184
- if hidden is None or hidden.get("quality_status") != "passed":
185
- raise TaskError("missing or invalid hidden group")
186
- items = hidden.get("items")
187
- if not isinstance(items, list) or len(items) != len(ids):
188
- raise TaskError("hidden item count mismatch")
189
- item_map = {}
190
- for item in items:
191
- if not isinstance(item, dict) or item.get("item_id") not in ids:
192
- raise TaskError("invalid hidden item")
193
- item_id = item["item_id"]
194
- if item_id in item_map or item.get("quality_status") not in ("retained", "retained_with_flag"):
195
- raise TaskError("duplicate or invalid hidden item")
196
- mean = finite(item.get("mean"), "mean")
197
- n = item.get("n")
198
- if n is not None and (type(n) is not int or n < 1):
199
- raise TaskError("invalid experimental replicate count")
200
- uncertainty_status = item.get("uncertainty_status")
201
- if uncertainty_status not in ("known", "unknown"):
202
- raise TaskError("invalid uncertainty status")
203
- uncertainty_model = group.get("uncertainty_model", "independent_replicate_mean")
204
- if uncertainty_model == "reported_sd_with_source_bounds":
205
- lo = finite(item.get("mean_lower"), "mean_lower")
206
- hi = finite(item.get("mean_upper"), "mean_upper")
207
- if lo > hi or not lo - 1e-12 <= mean <= hi + 1e-12:
208
- raise TaskError("inconsistent reported mean interval")
209
- if uncertainty_status == "unknown":
210
- if item.get("uncertainty_scale") is not None:
211
- raise TaskError("unknown uncertainty must be null")
212
- sem = None
213
- elif uncertainty_model == "optional_reported_uncertainty":
214
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
215
- if not item.get("uncertainty_basis"):
216
- raise TaskError("known uncertainty needs a stated source basis")
217
- elif uncertainty_model == "reported_sd_with_source_bounds":
218
- if n is None or n < 2:
219
- raise TaskError("reported SD needs repeat evidence")
220
- lower_sd = finite(item.get("sd_lower"), "sd_lower", 0)
221
- upper_sd = finite(item.get("sd_upper"), "sd_upper", 0)
222
- if lower_sd > upper_sd:
223
- raise TaskError("inconsistent reported SD interval")
224
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
225
- expected_scale = upper_sd + (hi - lo) / 2
226
- if not math.isclose(sem, expected_scale, rel_tol=1e-12, abs_tol=1e-12):
227
- raise TaskError("inconsistent source-bound uncertainty")
228
- elif uncertainty_model in ("independent_replicate_mean", "replicate_sd_upper_envelope"):
229
- sem = finite(item.get("uncertainty_scale"), "uncertainty_scale", 0)
230
- sd = finite(item.get("sample_sd"), "sample_sd", 0)
231
- if n is None or n < 2:
232
- raise TaskError("known SD needs repeat evidence")
233
- expected_sem = sd / math.sqrt(n) if uncertainty_model == "independent_replicate_mean" else sd
234
- if not math.isclose(sem, expected_sem, rel_tol=1e-12, abs_tol=1e-12):
235
- raise TaskError("inconsistent hidden uncertainty")
236
- else:
237
- raise TaskError("unsupported marginal uncertainty model")
238
- item_map[item_id] = {"mean": mean, "sem": sem, "uncertainty_status": uncertainty_status,
239
- "boundary_mean": item["mean"], "boundary_sem": item.get("uncertainty_scale")}
240
- stored = hidden.get("pairs")
241
- if not isinstance(stored, list) or len(stored) != len(ids) * (len(ids) - 1) // 2:
242
- raise TaskError("invalid hidden pair count")
243
- stored_map = {}
244
- for pair in stored:
245
- if not isinstance(pair, dict):
246
- raise TaskError("invalid hidden pair")
247
- key = (pair.get("item_i"), pair.get("item_j"))
248
- if key in stored_map:
249
- raise TaskError("duplicate hidden pair")
250
- stored_map[key] = pair
251
- pairs = []
252
- for item_i, item_j in itertools.combinations(ids, 2):
253
- left, right = item_map[item_i], item_map[item_j]
254
- difference = left["mean"] - right["mean"]
255
- known = left["uncertainty_status"] == right["uncertainty_status"] == "known"
256
- se = difference_se(left["sem"], right["sem"], method) if known else None
257
- tolerance = k * se + epsilon if known else epsilon
258
- comparable = int(pair_comparable(left["boundary_mean"], right["boundary_mean"],
259
- left["boundary_sem"] if known else None,
260
- right["boundary_sem"] if known else None,
261
- method, group.get("k", 2), group.get("epsilon", 0)))
262
- basis = "uncertainty_adjusted" if known else "reported_value"
263
- if not math.isfinite(difference) or not math.isfinite(tolerance) or (known and not math.isfinite(se)):
264
- raise TaskError("nonfinite pair arithmetic")
265
- pair = stored_map.get((item_i, item_j))
266
- if pair is None:
267
- raise TaskError("missing hidden pair")
268
- for name, expected, minimum in (("mean_difference", difference, None),):
269
- actual = finite(pair.get(name), name, minimum)
270
- if not math.isclose(actual, expected, rel_tol=1e-12, abs_tol=1e-10):
271
- raise TaskError("inconsistent hidden " + name)
272
- if known:
273
- if not math.isclose(finite(pair.get("difference_se"), "difference_se", 0), se,
274
- rel_tol=1e-12, abs_tol=1e-10):
275
- raise TaskError("inconsistent pair uncertainty")
276
- elif pair.get("difference_se") is not None:
277
- raise TaskError("unknown pair uncertainty must be null")
278
- if type(pair.get("comparable")) is not int or pair["comparable"] != comparable:
279
- raise TaskError("inconsistent comparable flag")
280
- if pair.get("uncertainty_status") != ("known" if known else "unknown"):
281
- raise TaskError("inconsistent pair uncertainty status")
282
- if pair.get("comparison_basis") != basis:
283
- raise TaskError("inconsistent comparison basis")
284
- pairs.append({"item_i": item_i, "item_j": item_j,
285
- "mean_difference": difference, "difference_se": se,
286
- "uncertainty_status": "known" if known else "unknown",
287
- "comparison_basis": basis, "comparable": comparable})
288
- total = sum(pair["comparable"] for pair in pairs)
289
- if total <= 0:
290
- raise TaskError("group has no scoreable pair under the declared rule")
291
- if type(hidden.get("comparable_pairs")) is not int or hidden["comparable_pairs"] != total:
292
- raise TaskError("inconsistent comparable pair count")
293
- validated.append({"group_id": gid, "item_ids": ids, "group_weight": group_weight,
294
- "pairs": pairs, "comparable_pairs": total})
295
- if set(hidden_map) != group_ids:
296
- raise TaskError("unknown hidden group")
297
- if not math.isclose(math.fsum(g["group_weight"] for g in validated), 1.0,
298
- rel_tol=0, abs_tol=1e-12):
299
- raise TaskError("group weights must sum to one")
300
- return validated
301
-
302
-
303
- def zero_report(contract, reason):
304
- return {"task_id": contract["task_id"], "task_version": contract["task_version"],
305
- "ranking_reward": 0.0, "submission_status": reason, "groups": [],
306
- "task_pair_accuracy": 0.0, "aggregation_zero_due_to_format": True,
307
- "official_reward": None, "review_status": "not_reviewed", "evaluation_status": "completed"}
308
-
309
-
310
- def grade_submission(contract, answers, submission, baseline):
311
- groups = validate_task(contract, answers)
312
- q = validate_baseline(contract, answers, baseline)
313
- if not isinstance(submission, dict):
314
- return zero_report(contract, "invalid_top_level_structure")
315
- if any(submission.get(key) != contract[key] for key in ("task_id", "task_version")):
316
- return zero_report(contract, "task_identity_mismatch")
317
- entries = submission.get("rankings")
318
- if not isinstance(entries, list):
319
- return zero_report(contract, "rankings_not_array")
320
- known = {g["group_id"] for g in groups}
321
- by_group = {}
322
- for entry in entries:
323
- if not isinstance(entry, dict) or not isinstance(entry.get("group_id"), str):
324
- return zero_report(contract, "unassignable_group_entry")
325
- if entry["group_id"] not in known:
326
- return zero_report(contract, "unknown_group_id")
327
- by_group.setdefault(entry["group_id"], []).append(entry)
328
- details = []
329
- for group in groups:
330
- gid = group["group_id"]
331
- matches = by_group.get(gid, [])
332
- item = {"group_id": gid, "group_weight": group["group_weight"],
333
- "group_pair_accuracy": 0.0, "status": "missing_group", "comparable_pairs": group["comparable_pairs"],
334
- "total_pairs": len(group["pairs"]), "item_count": len(group["item_ids"]),
335
- "coverage": group["comparable_pairs"] / len(group["pairs"])}
336
- if len(matches) > 1:
337
- item["status"] = "duplicate_group"
338
- elif len(matches) == 1:
339
- entry = matches[0]
340
- order = entry.get("order")
341
- if (not isinstance(order, list) or any(not isinstance(x, str) for x in order)
342
- or len(order) != len(group["item_ids"])
343
- or len(order) != len(set(order)) or set(order) != set(group["item_ids"])):
344
- item["status"] = "invalid_permutation"
345
- else:
346
- positions = {item_id: i for i, item_id in enumerate(order)}
347
- pair_details = []
348
- for pair in group["pairs"]:
349
- wrong = ((positions[pair["item_i"]] - positions[pair["item_j"]])
350
- * pair["mean_difference"]) < 0
351
- pair_details.append({**pair, "reversed": wrong,
352
- "counted_inversion": int(wrong and pair["comparable"] == 1)})
353
- lost = sum(pair["counted_inversion"] for pair in pair_details)
354
- total = group["comparable_pairs"]
355
- reward = 1.0 - lost / total
356
- item.update(status="valid", group_pair_accuracy=max(0.0, min(1.0, reward)),
357
- order=order, inversion_pairs=lost, correct_pairs=total-lost,
358
- comparable_pairs=total, total_pairs=len(pair_details),
359
- coverage=total/len(pair_details),
360
- reported_value_counted_pairs=sum(p["comparable"] for p in pair_details
361
- if p["comparison_basis"] == "reported_value"),
362
- uncertainty_adjusted_counted_pairs=sum(p["comparable"] for p in pair_details
363
- if p["comparison_basis"] == "uncertainty_adjusted"),
364
- all_reported_value_unequal_pairs=sum(p["mean_difference"] != 0 for p in pair_details),
365
- all_reported_value_correct_pairs=sum(p["mean_difference"] != 0 and not p["reversed"]
366
- for p in pair_details),
367
- pairs=pair_details)
368
- details.append(item)
369
- return {"task_id": contract["task_id"], "task_version": contract["task_version"],
370
- "task_pair_accuracy": aggregate_accuracy(details),
371
- "random_q95": q,
372
- "ranking_reward": reward_from_accuracy(aggregate_accuracy(details), q),
373
- "submission_status": "valid" if all(g["status"] == "valid" for g in details) else "group_errors",
374
- "groups": details, "official_reward": None,
375
- "review_status": "not_reviewed", "evaluation_status": "completed"}
376
-
377
-
378
- def read_regular(path, limit):
379
- try:
380
- fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK)
381
- try:
382
- info = os.fstat(fd)
383
- if not stat.S_ISREG(info.st_mode):
384
- raise SubmissionError("not_regular_file")
385
- if info.st_size > limit:
386
- raise SubmissionError("submission_too_large")
387
- with os.fdopen(fd, "rb", closefd=False) as stream:
388
- raw = stream.read(limit + 1)
389
- if len(raw) > limit:
390
- raise SubmissionError("submission_too_large")
391
- return raw
392
- finally:
393
- os.close(fd)
394
- except OSError as error:
395
- if error.errno in (errno.ENOENT, errno.ELOOP, errno.EISDIR, errno.ENOTDIR,
396
- errno.ENXIO, errno.EACCES, errno.EPERM):
397
- raise SubmissionError("missing_or_unreadable_submission") from error
398
- raise
399
-
400
-
401
- def grade_file(contract, answers, path, baseline):
402
- validate_task(contract, answers)
403
- q = validate_baseline(contract, answers, baseline)
404
- try:
405
- raw = read_regular(path, contract["max_submission_bytes"])
406
- submission = load_json(raw)
407
- except (SubmissionError, UnicodeError, json.JSONDecodeError, RecursionError) as error:
408
- report = zero_report(contract, str(error) or type(error).__name__)
409
- report["random_q95"] = q
410
- return report
411
- report = grade_submission(contract, answers, submission, baseline)
412
- report["random_q95"] = q
413
- report["submission_sha256"] = hashlib.sha256(raw).hexdigest()
414
- return report
415
-
416
-
417
- def dump(path, value):
418
- path.write_text(json.dumps(value, indent=2, allow_nan=False) + "\n", encoding="utf-8")
419
-
420
-
421
- def main():
422
- parser = argparse.ArgumentParser()
423
- parser.add_argument("--contract", type=Path, default=Path("/tests/contract.json"))
424
- parser.add_argument("--answers", type=Path, default=Path("/tests/answers.json"))
425
- parser.add_argument("--baseline", type=Path, default=Path("/tests/random_baseline.json"))
426
- parser.add_argument("--submission", type=Path, default=Path("/logs/artifacts/rankings.json"))
427
- parser.add_argument("--output-dir", type=Path, default=Path("/logs/verifier"))
428
- args = parser.parse_args()
429
- args.output_dir.mkdir(parents=True, exist_ok=True)
430
- for name in ("reward.txt", "reward.json", "report.json", "evaluator_error.json"):
431
- (args.output_dir / name).unlink(missing_ok=True)
432
- try:
433
- contract = load_json(args.contract.read_bytes())
434
- answers = load_json(args.answers.read_bytes())
435
- baseline = load_json(args.baseline.read_bytes())
436
- report = grade_file(contract, answers, args.submission, baseline)
437
- except (TaskError, SubmissionError, OSError, ValueError, KeyError, TypeError) as error:
438
- dump(args.output_dir / "evaluator_error.json", {"status": "task_error", "error": str(error)})
439
- raise SystemExit(2) from error
440
- dump(args.output_dir / "report.json", report)
441
- (args.output_dir / "reward.txt").write_text(str(report["ranking_reward"]) + "\n", encoding="utf-8")
442
- print(json.dumps({"ranking_reward": report["ranking_reward"],
443
- "submission_status": report["submission_status"],
444
- "task_pair_accuracy": report["task_pair_accuracy"],
445
- "random_q95": report["random_q95"],
446
- "aggregation_zero_due_to_format": report.get("aggregation_zero_due_to_format", False),
447
- "official_reward": None, "review_status": "not_reviewed", "evaluation_status": "completed"}))
448
-
449
-
450
- if __name__ == "__main__":
451
- main()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/random_baseline.json DELETED
@@ -1,54 +0,0 @@
1
- {
2
- "method": "uniform_complete_permutations",
3
- "generator": "numpy.random.PCG64",
4
- "numpy_version": "2.4.4",
5
- "quantile": 0.95,
6
- "quantile_method": "higher",
7
- "calibration_count": 200000,
8
- "validation_count": 200000,
9
- "calibration_seed": 9986557036737804553,
10
- "validation_seed": 1915439175653043376,
11
- "q": 0.7142857142857143,
12
- "task_id": "alkyl-iodide-nickel-difluoromethylation-additive-yield",
13
- "task_version": "1.0.0",
14
- "group_order": [
15
- "reaction_settings"
16
- ],
17
- "item_order": {
18
- "reaction_settings": [
19
- "exp_2dfe2f22dd32",
20
- "exp_34e91dee44f6",
21
- "exp_0003b83bcce3",
22
- "exp_595fa6ce8e2c",
23
- "exp_1da97213829d",
24
- "exp_2da4d7cbbf59",
25
- "exp_2191e72320e5",
26
- "exp_1638cbaa31ee",
27
- "exp_3e426daf35ae"
28
- ]
29
- },
30
- "calibration_code_sha256": "846e9c57f30b2c9fe9e310b7aeb965bd530e66c999eebca6aa6b70b546429891",
31
- "relations_sha256": "35e6f3cd119b548f696c2f5634f8be97293c43cf97c385575a72f2055d41399b",
32
- "weights_sha256": "c2272c0a862f11de35da66ec8be34349001ddebae43b85ab1dafb188f8ad7b30",
33
- "validation": {
34
- "task_pair_accuracy": {
35
- "mean": 0.4998055714285714,
36
- "mean_standard_error": 0.00030500088757490955
37
- },
38
- "task_spearman": {
39
- "mean": -0.0005092513195982953,
40
- "mean_standard_error": 0.0007910341459227771
41
- },
42
- "ranking_reward": {
43
- "mean": 0.010638500000000002,
44
- "mean_standard_error": 0.0001280075916438849
45
- },
46
- "zero_reward_fraction": 0.953065
47
- },
48
- "bindings": {
49
- "contract_sha256": "c39b5b1f4bce9cccd9c68c96036aca3ab696fa1f4458910be3a415b2268ece33",
50
- "answers_sha256": "0c3e957c4918c9cf3b769f9f96352585f6d29e17a05399ee0a504b38cf7a762f",
51
- "grader_sha256": "82c9d3b9c1fa5c2e2a39852e315df9e04714dee2ffa9eb64373014d8b4aeb381",
52
- "auxiliary_grader_sha256": "157f1e435b712cdaaa70ff7e8e9a372096898b1ea0b04dc2caa79a743c72cd34"
53
- }
54
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tasks/alkyl-iodide-nickel-difluoromethylation-additive-yield/tests/scoring_hashes.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "answers.json": "f1651c6f134d0dde2c9fad3554a7a7b0f41c61c7c2535555185e13cbad638aaf",
3
+ "contract.json": "7362f7a73b7e8fd77821cc4b662a7242bb69130540d9b406a77202dd28a8068e",
4
+ "grade.py": "0c25490ea9639fd4e73fde4741f4b4e02bdbd42962779348694e50e33a121550"
5
+ }
tasks/alkyne-bridged-ester-palladium-screen/environment/public/contract.json CHANGED
@@ -1,24 +1,20 @@
1
  {
2
  "task_id": "alkyne-bridged-ester-palladium-screen",
3
- "task_version": "1.0.0",
4
  "groups": [
5
  {
6
  "group_id": "palladium_settings",
7
  "label": "Palladium precursor settings",
8
  "item_ids": [
9
- "exp_1e79981891a3",
10
- "exp_1048291dee11",
11
  "exp_0887d89d8f86",
 
 
12
  "exp_3f91974be890",
13
- "exp_e58a8992674f",
14
- "exp_9c1a740ac5c4"
15
  ],
16
  "direction": "ascending",
17
  "group_weight": 1.0,
18
- "k": 2,
19
- "epsilon": 0,
20
- "uncertainty_model": "optional_reported_uncertainty",
21
- "difference_standard_error": "sem_sum_upper_bound",
22
  "endpoint": {
23
  "name": "GC yield of the specified benzoate",
24
  "unit": "%",
@@ -35,18 +31,9 @@
35
  "spearman_tie_method": "average",
36
  "spearman_undefined_policy": "task_error",
37
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
38
- "secondary_scoring_method": "equal_pair_inversion_random95",
39
- "ranking_reward_range": [
40
- 0,
41
- 1
42
- ],
43
- "random_quantile": 0.95,
44
- "unknown_uncertainty_policy": "use_reported_value",
45
- "pair_rule": "For each unordered pair, count it iff abs(mu_i-mu_j)>k*s_ij+epsilon for known uncertainty, or >epsilon for unknown uncertainty. Each counted pair has equal weight within its group. Unknown uncertainty is not zero uncertainty.",
46
- "ranking_reward_formula": "Validate finite 0.5<=q<1 and 0<=A<=1; if A==1 return 1; else if A<=q+1e-12 return 0; otherwise clamp((A-q)/(1-q),0,1). A is the fixed weighted sum of group pair accuracies; q is the whole-task random 95th percentile.",
47
  "format_errors": {
48
- "whole_task": "Core=-1; auxiliary=0 for malformed/oversized/non-UTF8 JSON, duplicate object keys, wrong/missing identity, invalid top-level/rankings, unknown or unassignable group.",
49
- "known_group": "Missing/duplicate known group or invalid item permutation: raw group Spearman=null, aggregation value=-1, group pair accuracy=0; preserve group weight and apply auxiliary threshold after aggregation."
50
  },
51
  "max_submission_bytes": 1048576,
52
  "minimum_conditions_per_group": 2,
@@ -63,5 +50,14 @@
63
  "report_path": "/logs/artifacts/report.md",
64
  "sources_path": "/logs/artifacts/sources.json",
65
  "analysis_path": "/logs/artifacts/analysis/",
66
- "forbidden_result_scope": "Experimental yields, rankings, selectivities or fitted response models for palladium-catalyzed alkyne-bridging C-C cleavage and esterification of unstrained alpha,beta-unsaturated carbonyl compounds with ortho-alkynylphenols, including the corresponding naphthyl aryl ester products and equivalent conversion/product-formation endpoints."
 
 
 
 
 
 
 
 
 
67
  }
 
1
  {
2
  "task_id": "alkyne-bridged-ester-palladium-screen",
3
+ "task_version": "1.1.0",
4
  "groups": [
5
  {
6
  "group_id": "palladium_settings",
7
  "label": "Palladium precursor settings",
8
  "item_ids": [
 
 
9
  "exp_0887d89d8f86",
10
+ "exp_1048291dee11",
11
+ "exp_1e79981891a3",
12
  "exp_3f91974be890",
13
+ "exp_9c1a740ac5c4",
14
+ "exp_e58a8992674f"
15
  ],
16
  "direction": "ascending",
17
  "group_weight": 1.0,
 
 
 
 
18
  "endpoint": {
19
  "name": "GC yield of the specified benzoate",
20
  "unit": "%",
 
31
  "spearman_tie_method": "average",
32
  "spearman_undefined_policy": "task_error",
33
  "spearman_aggregation": "Pearson correlation of true average ranks and submitted positions, separately within each group, then fixed group-weighted sum; do not pool groups or clip negative groups.",
 
 
 
 
 
 
 
 
 
34
  "format_errors": {
35
+ "whole_task": "Score -1 for malformed, oversized or non-UTF-8 JSON, duplicate object keys, incorrect task identity, invalid top-level structure, or an unknown or unassignable group.",
36
+ "known_group": "A missing or duplicate group or an invalid item permutation receives a group contribution of -1 before weighting. Its raw correlation is null."
37
  },
38
  "max_submission_bytes": 1048576,
39
  "minimum_conditions_per_group": 2,
 
50
  "report_path": "/logs/artifacts/report.md",
51
  "sources_path": "/logs/artifacts/sources.json",
52
  "analysis_path": "/logs/artifacts/analysis/",
53
+ "forbidden_result_scope": "Experimental yields, rankings, selectivities or fitted response models for palladium-catalyzed alkyne-bridging C-C cleavage and esterification of unstrained alpha,beta-unsaturated carbonyl compounds with ortho-alkynylphenols, including the corresponding naphthyl aryl ester products and equivalent conversion/product-formation endpoints.",
54
+ "supporting_artifact_review": {
55
+ "policy_stop_marker": "SPU_POLICY_STOP",
56
+ "required_files": [
57
+ "report.md",
58
+ "sources.json"
59
+ ],
60
+ "required_directory": "analysis",
61
+ "handling": "The verifier records file presence and declared stops separately in report.json. Reviewers use the run trajectory to assess compliance."
62
+ }
63
  }
tasks/alkyne-bridged-ester-palladium-screen/environment/public/experiment.md CHANGED
@@ -38,7 +38,7 @@ The screening experiments charge 0.1 mmol enone and 0.15 mmol alkynylphenol at 0
38
 
39
  ## Item settings
40
 
41
- Expand each item in targets.csv or protocol.json together with the common conditions. Every selected item is listed using the same fields.
42
 
43
  ## Source limitations relevant to the definition
44
 
 
38
 
39
  ## Item settings
40
 
41
+ Expand each item in targets.csv or protocol.json together with the common conditions.
42
 
43
  ## Source limitations relevant to the definition
44