FA-11846 / Search retrieval semantics / Open access

Synonym expansion changes a conjunction into a broad disjunction · case 01

Synonym expansion changes a conjunction into a broad disjunction.

Verified by executionVariant 1 · 7 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

Flattening synonym alternatives loses the required conjunction between query positions.

VERIFIED REPAIR

Require one document term from every alternative group.

Unsuccessful approach: Requiring every expanded token makes alternatives mandatory together.

Case contract

Return whether a token set satisfies all nonempty synonym groups. Each group is OR; groups are AND. Empty query matches; empty group cannot match.

Why this case matters

An offline deterministic retrieval model isolates this search contract from tokenization, storage, and network behavior.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(tokens, groups):
    return any(t in tokens for group in groups for t in group)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
term='car'+str(N)
check('one synonym per position', solve([term,'red'], [[term,'auto'],['red']]), True)
check('missing required position', solve([term], [[term,'auto'],['red']]), False)
check('alternate spelling', solve(['auto','red'], [[term,'auto'],['red']]), True)
check('no query', solve([], []), True)
check('empty alternative group', solve([term], [[]]), False)
check('unrelated document', solve(['blue'], [[term,'auto']]), False)
check('single exact term', solve([term], [[term]]), True)
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
one synonym per position	`True`	`True`	Passed
missing required position	`True`	`False`	Failed
alternate spelling	`True`	`True`	Passed
no query	`False`	`True`	Failed
empty alternative group	`False`	`False`	Passed
unrelated document	`False`	`False`	Passed
single exact term	`True`	`True`	Passed

SHA-256 / ffc4e070d6d5f8c3b6c8791dfd1bb10e2dc821d8bcaeb06fc43c25c91da5019c

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(tokens, groups):
    return all(t in tokens for group in groups for t in group)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
term='car'+str(N)
check('one synonym per position', solve([term,'red'], [[term,'auto'],['red']]), True)
check('missing required position', solve([term], [[term,'auto'],['red']]), False)
check('alternate spelling', solve(['auto','red'], [[term,'auto'],['red']]), True)
check('no query', solve([], []), True)
check('empty alternative group', solve([term], [[]]), False)
check('unrelated document', solve(['blue'], [[term,'auto']]), False)
check('single exact term', solve([term], [[term]]), True)
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
one synonym per position	`False`	`True`	Failed
missing required position	`False`	`False`	Passed
alternate spelling	`False`	`True`	Failed
no query	`True`	`True`	Passed
empty alternative group	`True`	`False`	Failed
unrelated document	`False`	`False`	Passed
single exact term	`True`	`True`	Passed

SHA-256 / cd8bb2db51370a5e58ce980fbfed5182e69c5f615f6a32388ebdd49cdcb6f9c1

3 / The verified repair

Exit 0

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

N = 1
observations = []
def solve(tokens, groups):
    return all(any(t in tokens for t in group) for group in groups)
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
term='car'+str(N)
check('one synonym per position', solve([term,'red'], [[term,'auto'],['red']]), True)
check('missing required position', solve([term], [[term,'auto'],['red']]), False)
check('alternate spelling', solve(['auto','red'], [[term,'auto'],['red']]), True)
check('no query', solve([], []), True)
check('empty alternative group', solve([term], [[]]), False)
check('unrelated document', solve(['blue'], [[term,'auto']]), False)
check('single exact term', solve([term], [[term]]), True)
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
one synonym per position	`True`	`True`	Passed
missing required position	`False`	`False`	Passed
alternate spelling	`True`	`True`	Passed
no query	`True`	`True`	Passed
empty alternative group	`False`	`False`	Passed
unrelated document	`False`	`False`	Passed
single exact term	`True`	`True`	Passed

SHA-256 / f3786ad3942ff52b7c51a41fc357825f73c69cb5ab875aab7415dc318eea621c

Verification & scope

Inputs are already tokenized or scored; this model makes no claim about production engine performance or linguistic analysis. 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 2026-09-29T14:38:51.552980+00:00.

Case digest / 0007d01cd62955986e3a0b252858a7c0e6d934b099088c985818af9ae16c6929