FA-98116 / Sequence batch packing / Open access

Continuation flags mislabel chunk boundaries · case 01

Continuation flags mislabel chunk boundaries.

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

ROOT CAUSE

The first chunk is marked as having a predecessor and every chunk is marked as continuing afterward.

THE FAILURE

The first chunk is marked as having a predecessor and every chunk is marked as continuing afterward.

Unsuccessful approach: Clearing the previous flag repairs the first chunk but still marks the final chunk as continuing afterward.

Case contract

Split into width-sized chunks and pair each with [continues_from_previous, continues_after], both booleans. Width is positive.

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(tokens, width):
    chunks=[tokens[i:i+width] for i in range(0,len(tokens),width)]
    return [[c,[i>0,True]] for i,c in enumerate(chunks)]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
data=list(range(N+1))
check('tail markers', solve(data,N), [[data[:N],[False,True]],[data[N:],[True,False]]])
check('single chunk', solve([N],N), [[[N],[False,False]]])
check('empty', solve([],N), [])
check('exact two chunks', solve(list(range(2*N)),N), [[list(range(N)),[False,True]],[list(range(N,2*N)),[True,False]]])
check('width larger', solve([1],N+1), [[[1],[False,False]]])
check('unit chunks', solve([4,5],1), [[[4],[False,True]],[[5],[True,False]]])
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
tail markers	`[[[0], [False, True]], [[1], [True, True]]]`	`[[[0], [False, True]], [[1], [True, False]]]`	Failed
single chunk	`[[[1], [False, True]]]`	`[[[1], [False, False]]]`	Failed
empty	`[]`	`[]`	Passed
exact two chunks	`[[[0], [False, True]], [[1], [True, True]]]`	`[[[0], [False, True]], [[1], [True, False]]]`	Failed
width larger	`[[[1], [False, True]]]`	`[[[1], [False, False]]]`	Failed
unit chunks	`[[[4], [False, True]], [[5], [True, True]]]`	`[[[4], [False, True]], [[5], [True, False]]]`	Failed

SHA-256 / 269157ebfb3351fce07c5643a2044e8d4924071e7c030aa76e997676eaa148f6

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(tokens, width):
    chunks=[tokens[i:i+width] for i in range(0,len(tokens),width)]
    return [[c,[False,True]] for i,c in enumerate(chunks)]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
data=list(range(N+1))
check('tail markers', solve(data,N), [[data[:N],[False,True]],[data[N:],[True,False]]])
check('single chunk', solve([N],N), [[[N],[False,False]]])
check('empty', solve([],N), [])
check('exact two chunks', solve(list(range(2*N)),N), [[list(range(N)),[False,True]],[list(range(N,2*N)),[True,False]]])
check('width larger', solve([1],N+1), [[[1],[False,False]]])
check('unit chunks', solve([4,5],1), [[[4],[False,True]],[[5],[True,False]]])
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
tail markers	`[[[0], [False, True]], [[1], [False, True]]]`	`[[[0], [False, True]], [[1], [True, False]]]`	Failed
single chunk	`[[[1], [False, True]]]`	`[[[1], [False, False]]]`	Failed
empty	`[]`	`[]`	Passed
exact two chunks	`[[[0], [False, True]], [[1], [False, True]]]`	`[[[0], [False, True]], [[1], [True, False]]]`	Failed
width larger	`[[[1], [False, True]]]`	`[[[1], [False, False]]]`	Failed
unit chunks	`[[[4], [False, True]], [[5], [False, True]]]`	`[[[4], [False, True]], [[5], [True, False]]]`	Failed

SHA-256 / 8c1bcf3711348ea37c908799c150f4e7b4e225f3aa5b617dc996ebd1b9d8117a

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 2026-09-30T16:48:09.322766+00:00.

Case digest / b8d52fa722eba691cb5a44e8381692b5dca57878645c728de91598a384a623e4