FA-100181 / Annotation span alignment / Open access
Overlapping annotation priority is ignored · case 01
Overlapping annotation priority is ignored.
ROOT CAUSE
The first covering span is selected without comparing annotation priorities.
THE FAILURE
The first covering span is selected without comparing annotation priorities.
Unsuccessful approach: Breaking equal-priority ties by later input order reverses the required stable tie rule.
Case contract
For each integer position in [0,length), choose covering span with highest priority; ties use earliest input; output one label per position or None.
Why this case matters
A small offline annotation transformation keeps span coordinates explicit when text or token boundaries change.
1 / The failure
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
N = 1
observations = []
def solve(spans, length):
return [next((label for s,e,label,p in spans if s<=x<e),None) for x in range(length)]
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('higher priority wins', solve([[1,4,'low',1],[2,5,'high',N+1]],6), [None,'low','high','high','high',None])
check('equal priority earlier wins', solve([[0,2,'first',N],[0,2,'second',N]],2), ['first','first'])
check('empty coverage', solve([],N), [None]*N)
check('zero output length', solve([[0,2,'x',1]],0), [])
check('nonoverlap labels', solve([[0,1,'a',1],[2,3,'b',1]],4), ['a',None,'b',None])
check('singleton highest', solve([[0,1,'a',2],[0,1,'b',3]],1), ['b'])
check('position at end excluded', solve([[0,1,'x',9]],2), ['x',None])
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 fixture | Actual | Expected | Outcome |
|---|---|---|---|
| higher priority wins | [None, 'low', 'low', 'low', 'high', None] | [None, 'low', 'high', 'high', 'high', None] | Failed |
| equal priority earlier wins | ['first', 'first'] | ['first', 'first'] | Passed |
| empty coverage | [None] | [None] | Passed |
| zero output length | [] | [] | Passed |
| nonoverlap labels | ['a', None, 'b', None] | ['a', None, 'b', None] | Passed |
| singleton highest | ['a'] | ['b'] | Failed |
| position at end excluded | ['x', None] | ['x', None] | Passed |
SHA-256 / 18c652773c2dafd769dc6621b5a7e36e7e6a52022bcb30aea020c429059636a5
2 / The unsuccessful fix
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
N = 1
observations = []
def solve(spans, length):
return [max(((p,i,label) for i,(s,e,label,p) in enumerate(spans) if s<=x<e),default=(0,0,None))[2] for x in range(length)]
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('higher priority wins', solve([[1,4,'low',1],[2,5,'high',N+1]],6), [None,'low','high','high','high',None])
check('equal priority earlier wins', solve([[0,2,'first',N],[0,2,'second',N]],2), ['first','first'])
check('empty coverage', solve([],N), [None]*N)
check('zero output length', solve([[0,2,'x',1]],0), [])
check('nonoverlap labels', solve([[0,1,'a',1],[2,3,'b',1]],4), ['a',None,'b',None])
check('singleton highest', solve([[0,1,'a',2],[0,1,'b',3]],1), ['b'])
check('position at end excluded', solve([[0,1,'x',9]],2), ['x',None])
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 fixture | Actual | Expected | Outcome |
|---|---|---|---|
| higher priority wins | [None, 'low', 'high', 'high', 'high', None] | [None, 'low', 'high', 'high', 'high', None] | Passed |
| equal priority earlier wins | ['second', 'second'] | ['first', 'first'] | Failed |
| empty coverage | [None] | [None] | Passed |
| zero output length | [] | [] | Passed |
| nonoverlap labels | ['a', None, 'b', None] | ['a', None, 'b', None] | Passed |
| singleton highest | ['b'] | ['b'] | Passed |
| position at end excluded | ['x', None] | ['x', None] | Passed |
SHA-256 / 2a0639a0671047132eee822cc6a956b17db86f4222d20d6a362275cfaff99d96
HELD IN THE MEMBER ARCHIVE
The verified repair and its recorded checks are member-only.
This mechanism has 7 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
Contracts use integer offsets, explicit edits, and supplied token boundaries; they do not define a tokenizer or production annotation format. 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 2026-09-30T16:48:32.646695+00:00.
Case digest / b87ea09cb48ed035e20c93f5750818b65ffba474f3237e4128979ce1cf5e6822