FA-98086 / Sequence batch packing / Open access

Packed positions count across segment boundaries · case 01

Packed positions count across segment boundaries.

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

ROOT CAUSE

Position indices are cumulative over a packed row rather than local to its segment.

THE FAILURE

Position indices are cumulative over a packed row rather than local to its segment.

Unsuccessful approach: One-based local positions still misalign every segment start.

Case contract

For each packed row of segment lengths, assign positions 0 through length-1 inside each segment, resetting at every boundary.

Why this case matters

This deterministic toy packing rule models a common boundary in preparing bounded training sequences.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(rows):
    return [[p for total in [sum(row)] for p in range(total)] for row in rows]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('two segment reset', solve([[N,2]]), [list(range(N))+[0,1]])
check('empty batch', solve([]), [])
check('empty row', solve([[]]), [[]])
check('zero segment', solve([[0,N]]), [list(range(N))])
check('several boundaries', solve([[1,3,2]]), [[0,0,1,2,0,1]])
check('separate rows', solve([[2],[N]]), [[0,1],list(range(N))])
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
two segment reset	`[[0, 1, 2]]`	`[[0, 0, 1]]`	Failed
empty batch	`[]`	`[]`	Passed
empty row	`[[]]`	`[[]]`	Passed
zero segment	`[[0]]`	`[[0]]`	Passed
several boundaries	`[[0, 1, 2, 3, 4, 5]]`	`[[0, 0, 1, 2, 0, 1]]`	Failed
separate rows	`[[0, 1], [0]]`	`[[0, 1], [0]]`	Passed

SHA-256 / 9329a192c133cd7da5538f94719caafd424ba6a50645dbc9157191ffabc44b2d

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(rows):
    return [[p for length in row for p in range(1,length+1)] for row in rows]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('two segment reset', solve([[N,2]]), [list(range(N))+[0,1]])
check('empty batch', solve([]), [])
check('empty row', solve([[]]), [[]])
check('zero segment', solve([[0,N]]), [list(range(N))])
check('several boundaries', solve([[1,3,2]]), [[0,0,1,2,0,1]])
check('separate rows', solve([[2],[N]]), [[0,1],list(range(N))])
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
two segment reset	`[[1, 1, 2]]`	`[[0, 0, 1]]`	Failed
empty batch	`[]`	`[]`	Passed
empty row	`[[]]`	`[[]]`	Passed
zero segment	`[[1]]`	`[[0]]`	Failed
several boundaries	`[[1, 1, 2, 3, 1, 2]]`	`[[0, 0, 1, 2, 0, 1]]`	Failed
separate rows	`[[1, 2], [1]]`	`[[0, 1], [0]]`	Failed

SHA-256 / db47bef7f113f83fb94aa1b431f77549766984fc66f0a54dee9db282a5e45bf1

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.

Verification & scope

Inputs are short integer-token lists under the stated toy contract; results do not claim compatibility with a specific training framework. 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 2026-09-30T16:48:09.322254+00:00.

Case digest / 28ddf6356d26496695ab8fe2fdfb06b6a38662823ddd8ed2409436a22230b46d