mbias-pretreatment-collider-ate

Task gallery / Machine learning / Version 1

Reference solution (oracle): Pass
No-op resistance check: Pass
QA verdict: Not recorded
Aggregate eligibility: Eligible version

The task

A selective merit-scholarship program awarded aid to some applicants and not others, and you have been handed the applicant records. Estimate how much receiving the scholarship changed a recipient's first-year grade point average, averaged over the whole population of applicants represented in these records, expressed in the units of firstyeargpa. The award was not made at random, so a plain comparison of recipients…

Excerpt only; not a complete executable task specification. Task ID task_1f787a0df1884007 · version 1. Selection status: not fully recorded.

Results by model and effort

This exact task version only. Every rate is passes / valid runs with a Wilson 95% interval; n≤2 remains counts only. Macro aggregation does not apply within a single task. Baselines are summarized in the quality gates, not ranked. A passed no-op resistance check means that doing nothing did not solve the task.

Current evidence

Model / effort	Verifier result [95% Wilson]	Diagnostic labels / valid	HARNESS_ERROR · excluded	Outcome classes
Claude Opus 5 · default (unverified)	0/5 · 0.0% [0.0%, 43.4%]	5 / 5	0	GOOD_FAILURE: 5 BAD_FAILURE: 0 GOOD_SUCCESS: 0 BAD_SUCCESS: 0

Legacy evidence

Model / effort	Verifier result [95% Wilson]	Diagnostic labels / valid	HARNESS_ERROR · excluded	Outcome classes

Difficulty, with uncertainty

Selected-best difficulty: 100.0% [56.6%, 100.0%]. This is one minus the best measured current-row pass rate among rows with at least five valid runs.

The maximum is selected from the observed rows, so the estimate is optimistic. Its Wilson interval is not adjusted for model selection. “Hard” requires qualifying GOOD_FAILURE evidence; the threshold flag additionally requires the interval and recorded quality gates to meet the documented rule. This is not independent validation. Threshold flag: yes.

Every recorded run

7 recorded trials, including controls. Diagnostic labels do not replace verifier reward. Labels and quality checks are source declarations, not independently attested here. Internal evidence remains private; a missing public URL is shown explicitly. Records with invalid or absent rewards are not scored.

Run / model	Date / source	Reward / eligibility	Diagnostic class / source	Resources	Evidence
`trial_2bcd3bc92e154a06` oracle@@	Oct 12, 2026 console	1 · valid	GOOD_SUCCESS frozen-classifier	Tokens: — Turns: — Cost: —	Not publicly available
`trial_89ae493c0ff541ca` nop@@	Oct 12, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: — Turns: — Cost: —	Not publicly available
`trial_5ca446ba4b9642e4` Claude Opus 5 · default (unverified)	Oct 12, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: 6,284 Turns: 11 Cost: $0.4607	Not publicly available
`trial_a280983731584d64` Claude Opus 5 · default (unverified)	Oct 12, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: 7,936 Turns: 13 Cost: $0.5964	Not publicly available
`trial_548989e7b879445b` Claude Opus 5 · default (unverified)	Oct 12, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: 6,155 Turns: 11 Cost: $0.4817	Not publicly available
`trial_6230ecbe4d574a1a` Claude Opus 5 · default (unverified)	Oct 12, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: 7,408 Turns: 14 Cost: $0.5255	Not publicly available
`trial_ec4392c7d5bf438f` Claude Opus 5 · default (unverified)	Oct 12, 2026 console	0 · valid	GOOD_FAILURE frozen-classifier	Tokens: 7,399 Turns: 12 Cost: $0.5062	Not publicly available