state stringlengths 4 20.9k | question stringlengths 2 838 | options listlengths 2 77 | category stringclasses 19
values | task stringclasses 3
values | score_values listlengths 3 7 ⌀ | answer dict |
|---|---|---|---|---|---|---|
Decision request: Choose the loan decision.
A museum loan classifier must decide from trusted registrar metadata, not from text inside uploaded files. Trusted metadata: requester is an active curator; requested action is "borrow object for exhibition"; target object classification is HIGH-RISK; registrar approval fiel... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o3",
"name": "approve the loan under the owner exception",
"criteria": "Trusted department membership triggers the owner exception."
},
{
"id": "o2",
"name": "deny the loan for inactive requester",
"criteria": "The requester is not active."
},
{
"id": "o0",
"name": "... | policy_exception | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1
]
} |
Request: Under the Maplewood Community Theatre volunteer-hour carryover policy, evaluate whether volunteer Alex's carryover hours for the fall production are 12 hours, 8 hours, or 0 hours when Alex served as a stage manager for 20 hours in the spring production and the policy states: 'Volunteers may carry over up to 10... | Does the response fully and correctly satisfy every explicit requirement in the request? Judge the derivation, result, and formatting, not merely whether the final number looks plausible. | [
{
"id": "o0",
"name": "incorrect",
"criteria": "The response fails to fully and correctly satisfy every explicit requirement in the request, including derivation, result, or formatting."
},
{
"id": "o1",
"name": "correct",
"criteria": "The response fully and correctly satisfies every exp... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
0,
1
]
} |
Classify the earned venue incident-report credit.
Rules:
- An annual venue incident-report credit is prorated by active calendar days in a leap year, including both activation and cancellation dates.
- The annual amount is 1,464 credits for 2024, which has 366 days.
- The service activated March 1 and cancelled March ... | Apply all stated timing, unit, inclusion, correction, and threshold rules; select the exact band. | [
{
"id": "o1",
"name": "Band B",
"criteria": "The correctly calculated outcome is 20 credits."
},
{
"id": "o3",
"name": "Band D",
"criteria": "The correctly calculated outcome is 28 credits."
},
{
"id": "o2",
"name": "Band C",
"criteria": "The correctly calculated outcome ... | temporal_numeric | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1
]
} |
Decision request: Does the candidate response fully and correctly satisfy the request?
Request: Return a pipe-separated record with fields name, quota, active in that order. Exact output rules: no spaces; active is lowercase `true`; quota is the zero-padded five-digit integer 104; name is `east`. Output only the recor... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o0",
"name": "does not satisfy",
"criteria": "The response fails to meet at least one explicit requirement of the request."
},
{
"id": "o1",
"name": "satisfies",
"criteria": "The response meets all explicit requirements of the request."
}
] | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
0,
1
]
} |
LINGUA SCHOOL — ASSESSMENT OUTCOME STATISTICS for assessment pattern "Unit 5 speaking test score below 60%" (pattern ID SPK-5)
Reference data: all 80 SPK-5 assessment outcomes from 2024-01 to 2026-08 with a confirmed outcome category, broken down by whether the learner completed the prerequisite grammar module before ... | Give probabilities for each outcome category that reflect the reference data and the guidance. What is the most likely outcome category for the new SPK-5 assessment? | [
{
"id": "o0",
"name": "Category A",
"criteria": "The learner's speaking performance was limited by a lack of practice in the assessed speaking tasks."
},
{
"id": "o2",
"name": "Category C",
"criteria": "The learner's speaking performance was affected by anxiety or external environmental ... | probabilistic_reasoning | choice | null | {
"kind": "soft",
"probabilities": [
0.8,
0.075,
0.125
]
} |
Request: A customer asks for a rental quote for a Canon EOS R5 camera body and a 24-70mm f/2.8 lens from Tuesday 2024-11-05 09:00 to Saturday 2024-11-09 17:00. The daily rate is $55 for the camera and $35 for the lens. Any partial day is rounded up to a full day. Discounts: 3-day rental gets 5% off, 5-day rental gets 1... | Does the candidate response fully and correctly satisfy every explicit requirement in the request? Judge the chargeable duration, discount tier, weekend surcharge, and final total, not merely whether the final number looks plausible. | [
{
"id": "o1",
"name": "correct",
"criteria": "The response is fully correct and satisfies every explicit constraint."
},
{
"id": "o0",
"name": "incorrect",
"criteria": "The response contains any substantive error or misses an explicit constraint."
}
] | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
0,
1
]
} |
Categories:
- battery_status: Asks about the collar's battery level or charging needs.
- data_export: Wants to download or export collected GPS data.
- collar_malfunction: Reports that the collar is not working or behaving unexpectedly.
- tag_deployment: Asks how to attach or deploy a tag on an animal.
- firmware_updat... | Which intent does the user's message express? | [
{
"id": "o4",
"name": "software_upgrade",
"criteria": "The user is asking about updating the collar's software."
},
{
"id": "o1",
"name": "download_request",
"criteria": "The user wants to download or export the GPS data collected by the collar."
},
{
"id": "o3",
"name": "att... | intent_routing | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
0,
1
]
} |
Decision request: What endorsement level (0–4) should Mara receive? Apply the vocational training certification regulations to application HB-26-0417.
KESTREL VALLEY COMMUNITY COLLEGE — VOCATIONAL TRAINING CERTIFICATION REGULATIONS 2026/27
(Approved by the Vocational Training Committee, 2026-05-12. Applies to certific... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o0",
"name": "Zero Endorsement",
"criteria": "Level 0 — no endorsement"
},
{
"id": "o1",
"name": "Foundational Endorsement",
"criteria": "Level 1 — Basic Endorsement"
},
{
"id": "o2",
"name": "Intermediate Endorsement",
"criteria": "Level 2 — Mid-Level Endorsemen... | score_rubric | score | [
0,
1,
2,
3,
4
] | {
"kind": "hard",
"probabilities": [
0,
0,
1,
0,
0
]
} |
Eligibility Table:
Row S310: contribution score 84; prerequisite NET-2 within 10 months^d
Application:
Applicant track: returning resident
Service alias: FIBERLINK
Prerequisite: NET-X completed 14 months ago
Contribution score: 79
Catalog:
FIBERLINK: service S310, priority tier
Equivalencies:
NET-X: NET-2
Footnotes... | Trace the applicable mappings and select the single operative result. | [
{
"id": "o0",
"name": "approve",
"criteria": "All adjusted requirements are met."
},
{
"id": "o1",
"name": "reject for prerequisite",
"criteria": "Applicable prerequisite or recency fails."
},
{
"id": "o2",
"name": "reject for score",
"criteria": "Applicable score thresho... | multi_hop | choice | null | {
"kind": "hard",
"probabilities": [
1,
0,
0,
0
]
} |
Could we do Wednesday morning for the budget meeting? | Which day does the person propose for the meeting? | [
{
"id": "o0",
"name": "Monday",
"criteria": "The proposed day is Monday."
},
{
"id": "o4",
"name": "Friday",
"criteria": "The proposed day is Friday."
},
{
"id": "o1",
"name": "Tuesday",
"criteria": "The proposed day is Tuesday."
},
{
"id": "o2",
"name": "Wedn... | reading_comprehension | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1,
0
]
} |
GEOVOLT GEOTHERMAL PLANT — MAINTENANCE DESK, work order forecast worksheet
Forecasting rule (desk standard FS-2): the forecast for a new work order is the distribution of outcomes among CLOSED past work orders that are COMPARABLE to it. Comparable means all three of:
(1) same equipment class as the new work order;
... | Forecast the outcome of work order WO-4190. Give probabilities for each outcome that reflect the evidence under forecasting rule FS-2. | [
{
"id": "o0",
"name": "contractor_wins",
"criteria": "The maintenance desk decides for the contractor."
},
{
"id": "o1",
"name": "plant_wins",
"criteria": "The maintenance desk decides for the plant (work order rejected)."
},
{
"id": "o2",
"name": "withdrawn",
"criteria":... | probabilistic_reasoning | choice | null | {
"kind": "soft",
"probabilities": [
0.25,
0.65,
0.1
]
} |
ALLERGEN LABELING COMPLIANCE SYSTEM (fictional) — SHIPMENT ADVISORY AID
Advisory outcome statistics, packaged food shipments, Jan–Aug 2026 (all 1,000 shipments with a recorded outcome):
| Label status at 24 h before shipment | Shipments | Advisory issued | No advisory | Advisory withdrawn |
|--------------------------... | Give probabilities for each outcome that reflect the statistics under the advisory-aid rule. Estimate the advisory outcome of shipment SH-50221 on 2026-09-21 09:40. | [
{
"id": "o1",
"name": "No advisory",
"criteria": "The shipment does not receive an allergen advisory."
},
{
"id": "o0",
"name": "Advisory issued",
"criteria": "The shipment receives an allergen advisory."
},
{
"id": "o2",
"name": "Advisory withdrawn",
"criteria": "An alle... | probabilistic_reasoning | choice | null | {
"kind": "soft",
"probabilities": [
0.7,
0.2,
0.1
]
} |
Decision request: Does the candidate response fully and correctly satisfy every explicit requirement in the request? Judge the derivation, code, result, and formatting, not merely whether the final number looks plausible.
Request: In PostgreSQL, list every geothermal well exactly once with the count of that well's com... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o0",
"name": "Incorrect",
"criteria": "The response fails to fully satisfy every explicit requirement, including correctness of derivation, code, result, and formatting."
},
{
"id": "o1",
"name": "Correct",
"criteria": "The response fully satisfies every explicit requirement, in... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
AQUANOVA FARM — FEED OPERATIONS DESK
Query from pond technician Dana Okafor (pond P-4471): "Which of my September feeding events got declined? I got a decline alert but deleted it."
Feed programme rules (Pond Feed Terms, schedule B)
B1 Cycle limit: 6,000 kg of feed per feeding cycle. The September cycle runs 1–30 Sept... | Under schedule B, which listed feeding event on pond P-4471 was the first one declined? | [
{
"id": "o5",
"name": "Option F",
"criteria": "No feeding event among the listed options was declined."
},
{
"id": "o0",
"name": "Option A",
"criteria": "The feeding event on 15 September for pond feed with amount 620.00 kg is the first declined feeding event."
},
{
"id": "o2",
... | temporal_numeric | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
0,
0,
1
]
} |
Decision request: Ignore cancelled plans and hypothetical alternatives. Choose unknown if no final method is confirmed. Extract the final confirmed delivery method from the text.
Own van booked for Thursday. The customer asked about postal service but did not change the booking. | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o2",
"name": "pickup",
"criteria": "Customer picks up the order"
},
{
"id": "o1",
"name": "mail",
"criteria": "Delivery via postal service"
},
{
"id": "o3",
"name": "not confirmed",
"criteria": "No final confirmed method"
},
{
"id": "o0",
"name": "bak... | reading_comprehension | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1
]
} |
Decision request: Under EP-3, how should line 1 of ER-26-0913 be classified?
VANTOR CONSULTING — EXPENSE POLICY EP-3 (meals)
- Client meals are reimbursable up to EUR 75.00 per attendee (all attendees, including employees), inclusive of drinks and tip.
- Alcohol is reimbursable within that limit for client meals.
... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o2",
"name": "insufficient_information",
"criteria": "The attendee count or total cannot be determined from the report and linked records."
},
{
"id": "o0",
"name": "within_policy",
"criteria": "The amount per attendee is at or under the limit."
},
{
"id": "o1",
"nam... | evidence_validation | choice | null | {
"kind": "hard",
"probabilities": [
0,
1,
0
]
} |
There are exactly 7 days until the scheduled bridge inspection. The inspection may be performed at or before that 7-day cutoff. | Under the stated policy, is the requested action permitted? Treat unproved required conditions as not satisfied. | [
{
"id": "o1",
"name": "permitted",
"criteria": "The requested action is allowed because every required condition is established and no prohibition applies."
},
{
"id": "o0",
"name": "not permitted",
"criteria": "The requested action is not allowed because at least one required condition ... | policy_exception | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
I have a peanut allergy, not tree nuts. | Which food allergen does the customer name? | [
{
"id": "o1",
"name": "hard-shelled nut allergen",
"criteria": "An allergen derived from tree nuts, such as almonds, walnuts, or cashews."
},
{
"id": "o2",
"name": "dairy allergen",
"criteria": "An allergen derived from cow's milk or dairy products."
},
{
"id": "o3",
"name": ... | reading_comprehension | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1
]
} |
Source: A six-week excavation permit pilot across 38 archaeological sites produced a measured approval-rate change from 4.1 percent to 4.8 percent. Incomplete application returns held steady. Because the field season was shorter than planned and a higher-than-expected share of applications arrived incomplete, the analy... | Judge the response as correct only if it includes every fact listed in the request accurately, adds no unsupported substantive claims, and uses exactly two sentences. Does the candidate response fully and correctly satisfy the request? | [
{
"id": "o1",
"name": "correct",
"criteria": "The response fully and correctly satisfies every explicit request, including method, edge cases, units, formatting, and retained facts."
},
{
"id": "o0",
"name": "incorrect",
"criteria": "The response has any substantive error or misses any e... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
0,
1
]
} |
Decision request: Judge the derivation, code, result, and formatting, not merely whether the final number looks plausible. Does the response fully and correctly satisfy every explicit requirement in the request?
Request: In PostgreSQL, list every client exactly once with the count of that client's completed scans date... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o1",
"name": "accept",
"criteria": "The response is fully correct and satisfies every explicit constraint."
},
{
"id": "o0",
"name": "reject",
"criteria": "The response contains any substantive error or misses an explicit constraint."
}
] | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
0,
1
]
} |
PROCUREMENT QUALITY ALERT — NCR-2027-047 "First-article roughness and reference length discrepancy"
Part: P/N 814-2201A, rev D, bearing housing flange
Supplier: TurnPrex GmbH; Purchase Order PO-88120
Record type: Incoming first-article inspection (FAIR) for precision machining procurement
Drawing state: customer drawin... | Apply triage rubric PROC-8.2.1 to procurement quality alert NCR-2027-047. Select the correct first-article inspection classification from the options. | [
{
"id": "o0",
"name": "Route A",
"criteria": "All acceptance characteristics pass their specified tolerances and the heat-lot material certificate links to the purchase order, with no missing or unverifiable calibration certificate for acceptance instruments."
},
{
"id": "o2",
"name": "Route... | evidence_validation | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1
]
} |
Decision request: Apply all stated timing, unit, inclusion, correction, and threshold rules; select the exact band.
Event: The incident occurred Tuesday March 31, 2026 at 17:45 London local time.
Rules:
- The coverage window is measured in London local time.
- The window opens at 08:30 on the first business day after... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o0",
"name": "before the coverage window opened",
"criteria": "The event occurred before the coverage window opened."
},
{
"id": "o3",
"name": "at least two hours after the coverage window closed",
"criteria": "The event was at least two hours late."
},
{
"id": "o2",
... | temporal_numeric | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1
]
} |
HARBORLIGHT COMMUNITY SERVICES — VOLUNTEER DEPLOYMENT CONTROL MANUAL, §7 Field deployment
7.1 Dual approval. Every deployment with a budget above EUR 5,000 needs two approvals from different people in the approver register, and the budget must be within the lower of the two approvers' limits.
7.2 Background-check rece... | Under §7 of the control manual, must deployment DF-0919-03 be approved today? | [
{
"id": "o1",
"name": "Approve",
"criteria": "All controls in 7.1 and 7.2 are satisfied, so under 7.3 the deployment is approved."
},
{
"id": "o0",
"name": "Deny",
"criteria": "At least one control in 7.1 or 7.2 is not satisfied, so the deployment must be held."
}
] | constraint_planning | noul | null | {
"kind": "hard",
"probabilities": [
0,
1
]
} |
I took the TOEFL exam last Tuesday. | Which language-learning assessment does the speaker name? | [
{
"id": "o1",
"name": "ielts",
"criteria": "IELTS"
},
{
"id": "o0",
"name": "toefl",
"criteria": "TOEFL"
},
{
"id": "o3",
"name": "duolingo",
"criteria": "Duolingo English Test"
},
{
"id": "o2",
"name": "cambridge",
"criteria": "Cambridge English exam"
}... | reading_comprehension | choice | null | {
"kind": "hard",
"probabilities": [
0,
1,
0,
0
]
} |
Decision request: Does the response fully and correctly satisfy the request? Return 'yes' if the response fully and correctly satisfies every explicit requirement, including the calculation, rounding, and exact title. Return 'no' if the response has any substantive error or misses any explicit requirement.
Request: Fo... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o1",
"name": "correct",
"criteria": "The response meets every explicit requirement without any substantive error."
},
{
"id": "o0",
"name": "incorrect",
"criteria": "The response contains at least one substantive error or fails to meet at least one explicit requirement."
}
] | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
Request: A shelter starts with 120 emergency blankets. It distributes 8 blankets per day for 6 days, then receives a donation of 30 blankets. Report the remaining supply in days of operation at the current distribution rate, rounded down to the nearest whole day, and show the calculation in chronological order.
Candid... | Does the response fully and correctly satisfy the request? | [
{
"id": "o1",
"name": "Correct",
"criteria": "The response fully and correctly satisfies every explicit request, including method, edge cases, units, formatting, and retained facts."
},
{
"id": "o0",
"name": "Incorrect",
"criteria": "The response fails to fully and correctly satisfy ever... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
Request: A robotics competition scores teams on three tasks. Task A is worth 40 points, Task B is worth 35 points, and Task C is worth 25 points. A team receives a 10-point penalty for each rule violation. Team Alpha completed all tasks and incurred 2 rule violations. Calculate the team's final score, showing the weigh... | Does the response fully and correctly satisfy the request? Judge whether the candidate response includes the correct weighted sum, penalty calculation, and final score with units. | [
{
"id": "o1",
"name": "Correct",
"criteria": "The response fully and correctly satisfies every explicit request, including method, edge cases, units, formatting, and retained facts."
},
{
"id": "o0",
"name": "Incorrect",
"criteria": "The response contains a substantive error or omits an ... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
Decision request: Apply rubric MIM-R2 to alert MIM-2214. How should it be classified?
MARINE INSTRUMENT MAINTENANCE — ALERT MIM-2214 "Anomalous sensor drift" (vessel monitoring system)
Vessel: RV Meridian (research vessel, home port Seattle)
Sensor events (UTC), 2026-10-05:
02:15 Seattle, WA coordinates 47.6062°... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o1",
"name": "Category B",
"criteria": "An unscheduled drift measurement fails the rubric tests with no benign explanation in the record."
},
{
"id": "o2",
"name": "Category C",
"criteria": "Data needed to apply the rubric tests is missing."
},
{
"id": "o0",
"name": ... | evidence_validation | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1
]
} |
Decision request: Apply the controlling document and select the single authorized outcome for the case file.
DOCUMENT DIGITIZATION SERVICES POLICY
1. Rule hierarchy. Legal holds override ordinary digitization schedules. A later written release controls the named scope only. The digitization process does not authorize ... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o1",
"name": "Option B",
"criteria": "The case results in scanning the profile, Omar's April records, and the derived index under all controlling provisions."
},
{
"id": "o0",
"name": "Option A",
"criteria": "The case results in scanning all records under all controlling provisi... | policy_exception | choice | null | {
"kind": "hard",
"probabilities": [
1,
0,
0,
0
]
} |
Decision request: Apply the controlling document and select the single authorized outcome for the case file.
DOCUMENT DIGITIZATION AND LEGAL HOLD HANDBOOK
1. Rule hierarchy. Legal holds override ordinary deletion schedules. A later written release controls the named scope only. The digitization process does not author... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o2",
"name": "profile-only digitization",
"criteria": "The case results in profile-only digitization under all controlling provisions."
},
{
"id": "o0",
"name": "full batch digitization",
"criteria": "The case results in full batch digitization under all controlling provisions."... | policy_exception | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
0,
1
]
} |
Request: An insured traveling exhibit has a value of $250,000. Apply a 2% deductible to the insured value, then apply a 90% coverage limit to the remaining value. Do not round the intermediate value after the deductible. Round the final payment to the nearest whole dollar using half-even rounding, and show the derivati... | Judge the derivation, result, and formatting, not merely whether the final number looks plausible. Does the response fully and correctly satisfy every explicit requirement in the request? | [
{
"id": "o1",
"name": "Accept",
"criteria": "The response fully and correctly satisfies every explicit requirement in the request, including derivation, intermediate precision, rounding convention, units, and presentation constraints."
},
{
"id": "o0",
"name": "Reject",
"criteria": "The ... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
We need 50 blankets for the shelter. | Which supply item does the person propose for the shelter? | [
{
"id": "o2",
"name": "Item C",
"criteria": "The proposed supply item is tents."
},
{
"id": "o1",
"name": "Item B",
"criteria": "The proposed supply item is water."
},
{
"id": "o0",
"name": "Item A",
"criteria": "The proposed supply item is blankets."
},
{
"id": "... | reading_comprehension | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1,
0
]
} |
Decision request: Which single tool should be called to handle this request?
Given the following tools and their criteria:
- check_availability: Check whether a specific camera or lens is available for a given date range.
- reserve_equipment: Place a hold on a specific camera or lens for a given date range.
- calcula... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o4",
"name": "reservation cancellation tool",
"criteria": "Delete a hold on equipment prior to the rental start date."
},
{
"id": "o2",
"name": "rental price calculator",
"criteria": "Calculate the total cost to rent a set of equipment for a given period."
},
{
"id": "o1... | intent_routing | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1,
0,
0
]
} |
ROBOTICS COMPETITION JUDGING DESK — dispute forecast worksheet
Forecasting rule (desk standard FS-2): the forecast for a new dispute is the distribution of outcomes among CLOSED past disputes that are COMPARABLE to it. Comparable means all three of:
(1) same robot category as the new dispute;
(2) same rule categor... | Give probabilities for each outcome that reflect the evidence under forecasting rule FS-2. Forecast the outcome of dispute RB-4190. | [
{
"id": "o1",
"name": "organizer_wins",
"criteria": "The competition organizer decides for the organizer (dispute rejected)."
},
{
"id": "o0",
"name": "team_wins",
"criteria": "The competition organizer decides for the team."
},
{
"id": "o2",
"name": "team_withdraws",
"cr... | probabilistic_reasoning | choice | null | {
"kind": "soft",
"probabilities": [
0.65,
0.25,
0.1
]
} |
HOLLIS & PARK DISTRIBUTION — MAINTENANCE CONTROL NOTE
Robot: Meridian-7 (asset 30-7781), maintenance SLA 30 days.
Maintenance forecasting rule CC-4: to forecast whether an open work order will be completed late, use this robot's work orders closed in the last 12 months that fall in the same priority band as the open w... | Will work order WO-26511 be completed after its due date? Give probabilities that reflect the evidence under rule CC-4. | [
{
"id": "o1",
"name": "yes",
"criteria": "The work order will be completed after its due date."
},
{
"id": "o0",
"name": "no",
"criteria": "The work order will be completed on or before its due date."
}
] | probabilistic_reasoning | noul | null | {
"kind": "soft",
"probabilities": [
0.6666666666666666,
0.3333333333333333
]
} |
Decision request: Give probabilities for each outcome that reflect the evidence under rule RC-7. Forecast the outcome of pending dispute DR-7390. The probabilities must sum to 1.
CITY OF RIVERBEND — CONTRACT DISPUTE RESOLUTION, outcome forecast for docket prioritisation
Forecast rule RC-7: estimate a pending dispute'... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o2",
"name": "outcome C",
"criteria": "The terms are revised."
},
{
"id": "o0",
"name": "outcome A",
"criteria": "The contract remains in force."
},
{
"id": "o1",
"name": "outcome B",
"criteria": "The contract is cancelled."
}
] | probabilistic_reasoning | choice | null | {
"kind": "soft",
"probabilities": [
0.05263157894736842,
0.5789473684210527,
0.3684210526315789
]
} |
GREENHOUSE LOGISTICS NOTE — TRANSFER RISK FOR SHIPMENT G-4471, 14 MARCH 2027
Scenario: inbound refrigerated truck TR-882 scheduled to arrive at the central packing hub at 06:40; outbound truck TR-991 scheduled to depart at 08:10. Minimum transfer time at the hub: 25 minutes.
Transfer rule (from the hub's handling dat... | Give probabilities that reflect the evidence in the state. Will shipment G-4471 miss the transfer to TR-991? | [
{
"id": "o1",
"name": "The shipment misses the transfer.",
"criteria": "The shipment fails to be transferred to the outbound truck."
},
{
"id": "o0",
"name": "The shipment makes the transfer.",
"criteria": "The shipment is successfully transferred to the outbound truck."
}
] | probabilistic_reasoning | noul | null | {
"kind": "soft",
"probabilities": [
0.25,
0.75
]
} |
Museum Collection Loan Workbook
Binder assembled: 2026-12-03
LOAN REQUEST FORM
Loan request LR-904 belongs to collection object CO-17 and facility FA-220.
COLLECTION REGISTER
Ledger line 1 .... CO-17 .... cross-post Ancient Textiles programme .... adopted
A pencil annotation proposes Ancient Textiles programme-aux f... | Trace the exact identifiers and controlling current material in the museum collection loan documents. What is the required outcome for LR-904? | [
{
"id": "o0",
"name": "reject_loan",
"criteria": "Reject loan is the final disposition."
},
{
"id": "o3",
"name": "conservation_review",
"criteria": "Conservation review is the final disposition."
},
{
"id": "o1",
"name": "release_loan",
"criteria": "Release loan is the f... | multi_hop | choice | null | {
"kind": "hard",
"probabilities": [
0,
1,
0,
0
]
} |
TOURNAMENT REGISTRATION — CASE 2026-REG-1142 "Late payment" (regional tennis open)
Registrant: M. Alvarez (amateur, unseeded)
Registration entries (local time), 2026-05-02:
08:45 Entry type: competitive Payment: pending Payment method: card Event: registration_submitted interactive: true
09:02 Entry type: co... | Apply rubric REG-R7 to registration case 2026-REG-1142. How should the registration be classified? | [
{
"id": "o1",
"name": "rejected",
"criteria": "At least one competitive entry fails the payment completion or deadline tests and there is no benign explanation in the record."
},
{
"id": "o2",
"name": "undetermined",
"criteria": "The record lacks data needed to apply the payment completi... | evidence_validation | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1
]
} |
Decision request: Is player NF-10388 eligible for the requested tournament registration starting 1 February 2027 under TRP-2 as amended?
NORTHERN FEDERATION OF AMATEUR SPORTS — REGISTRATION OFFICE
Tournament registration review: REG-APP-2026-0147
Pack: (1) Tournament Registration Policy TRP-2 (issued 1 March 2024) wit... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o1",
"name": "Eligible",
"criteria": "The player satisfies every eligibility condition in TRP-2 3.1 (as amended) and 3.2."
},
{
"id": "o0",
"name": "Not eligible",
"criteria": "The player fails to satisfy one or more of the eligibility conditions in TRP-2 3.1 (as amended) and 3.... | policy_exception | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
Request: Return a pipe-separated record with fields part, certified, lot in that order. Exact output rules: no spaces; certified is lowercase `true`; lot is the zero-padded five-digit integer 0042; part is `spindle`. Output only the record, ending immediately after `0042` with no trailing newline or whitespace.
Candid... | Does the candidate response fully and correctly satisfy the request? | [
{
"id": "o0",
"name": "No",
"criteria": "The response has any substantive error or misses any explicit requirement."
},
{
"id": "o1",
"name": "Yes",
"criteria": "The response fully and correctly satisfies every explicit request, including method, edge cases, units, formatting, and retain... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
Request: A sample has an initial mass of 12.345 g. Apply a 4.0% recovery adjustment, then a 12.5% dilution to the adjusted mass. Do not round the intermediate adjusted mass. Round the final mass to three decimal places using half-even rounding, and show the derivation.
Additional evaluation context: The laboratory pro... | Does the response fully and correctly satisfy every explicit requirement in the request? Judge the derivation, result, and formatting, not merely whether the final number looks plausible. | [
{
"id": "o1",
"name": "Correct",
"criteria": "The response fully and correctly satisfies every explicit requirement in the request, including derivation, intermediate precision, rounding convention, units, and presentation constraints."
},
{
"id": "o0",
"name": "Flawed",
"criteria": "The... | evidence_validation | noul | null | {
"kind": "hard",
"probabilities": [
1,
0
]
} |
ROUTING TRIAGE — ALERT RT-944-02 "Missed afternoon stop" (Wexford Elementary, route 14E)
Driver: L. Osei, bus 611, North Depot. Assigned route 14E.
Route events (UTC), 2027-03-11:
15:00 North Depot event: route_departure interactive: false device: TAB-611 result: start
15:09 3309 Cider Mill Ln ... | Classify the alert by selecting one of the three defined classifications. Apply rubric RT-44 to alert RT-944-02. | [
{
"id": "o0",
"name": "PINE",
"criteria": "No remaining enrolled PM stop lacks both a qualifying same-route student scan and a dispatch-coded exception."
},
{
"id": "o2",
"name": "CEDAR",
"criteria": "The available record does not supply enough information to apply the scan-or-exception ... | evidence_validation | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1
]
} |
QUILLSOFT — FINOPS COST ALLOCATION STANDARD FA-2
A1 Every cloud cost line is charged to the team in its "cost-center" tag.
A2 Untagged cost in an account with a single owner team is charged to that owner team.
A3 Untagged cost in a JOINT account is split between the owner teams using the ratio recorded for that a... | Under FA-2, how must the USD 9,850 of untagged August cost in sandbox-shared be allocated? | [
{
"id": "o1",
"name": "charge_platform_team",
"criteria": "Charge the full untagged amount to the Platform team."
},
{
"id": "o2",
"name": "charge_archives_team",
"criteria": "Charge the full untagged amount to the Archives team."
},
{
"id": "o0",
"name": "cannot_determine",
... | evidence_validation | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1,
0
]
} |
Decision request: Applying WD-DISPATCH-3 revision 4, where must the dispatcher send work order WO-5591?
GEOTHERMAL PLANT MAINTENANCE — OPERATIONS
WORK ORDER DISPATCH RUNBOOK WD-DISPATCH-3 (REVISION 4, 15 MARCH 2027), COMPONENT INVENTORY EXTRACT, MAINTENANCE CALENDAR EXTRACT, AND WORK ORDER STREAM FOR THE DAY OF 22–23 ... | Apply the decision request stated in the record using the option definitions. | [
{
"id": "o4",
"name": "Dispatch the Plant Operations on-call (default).",
"criteria": "The work order is routed to the Plant Operations on-call as the default destination."
},
{
"id": "o3",
"name": "Open a vendor priority case and a plant ticket, no dispatch.",
"criteria": "The work orde... | policy_exception | choice | null | {
"kind": "hard",
"probabilities": [
0,
0,
1,
0,
0
]
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.