FAILURE MAP
← Case archive

FA-100136 / Preference pair construction / Open access

Preference pairs: pair text budget is consumed by the first endpoint · case 01

pair text budget is consumed by the first endpoint.

Verified by executionVariant 1 · 6 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

A shared truncation cursor leaves the rejected endpoint with only unused remainder.

THE FAILURE

A shared truncation cursor leaves the rejected endpoint with only unused remainder.

Unsuccessful approach: Splitting the total budget in half truncates short chosen texts unnecessarily.

Case contract

Rows are [chosen,rejected]. Truncate each endpoint independently to the first limit characters; limit is nonnegative.

Why this case matters

A small deterministic preference-data transform makes pair semantics inspectable without a training framework.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(rows, limit):
    return [[(r[0]+r[1])[:limit],(r[0]+r[1])[limit:2*limit]] for r in rows]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('full budget each',solve([['chosen','rejected']],4),[['chos','reje']])
check('zero budget',solve([['abc','xyz']],0),[['','']])
check('empty rows',solve([],N),[])
check('short text',solve([['a','xy']],5),[['a','xy']])
check('empty rejected',solve([['abc','']],2),[['ab','']])
check('exact boundary',solve([['abcd','wxyz']],4),[['abcd','wxyz']])
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
full budget each[['chos', 'enre']][['chos', 'reje']]Failed
zero budget[['', '']][['', '']]Passed
empty rows[][]Passed
short text[['axy', '']][['a', 'xy']]Failed
empty rejected[['ab', 'c']][['ab', '']]Failed
exact boundary[['abcd', 'wxyz']][['abcd', 'wxyz']]Passed

SHA-256 / 38983b3beaf43730140cff8a2ac37f22fef5f189ef15dd960686d771ae585a2c

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(rows, limit):
    return [[r[0][:limit//2],r[1][:limit//2]] for r in rows]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('full budget each',solve([['chosen','rejected']],4),[['chos','reje']])
check('zero budget',solve([['abc','xyz']],0),[['','']])
check('empty rows',solve([],N),[])
check('short text',solve([['a','xy']],5),[['a','xy']])
check('empty rejected',solve([['abc','']],2),[['ab','']])
check('exact boundary',solve([['abcd','wxyz']],4),[['abcd','wxyz']])
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
full budget each[['ch', 're']][['chos', 'reje']]Failed
zero budget[['', '']][['', '']]Passed
empty rows[][]Passed
short text[['a', 'xy']][['a', 'xy']]Passed
empty rejected[['a', '']][['ab', '']]Failed
exact boundary[['ab', 'wx']][['abcd', 'wxyz']]Failed

SHA-256 / dc5f1c6d2e064d2be7e01907b5e142e7000af0918a128c09f8e3e6b24638e208

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 6 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Sign in to the archive ↗

Verification & scope

These are explicit toy record contracts; they do not claim compatibility with a trainer or improved model behavior. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-30T16:48:32.170476+00:00.

Case digest / c447f95085a7ddab47c0db5a791fb059ff20bc2b473d8942b7d3240594256805