apiVersion: "science-ladder/v1"
kind: "ChallengeCandidate"
id: "quiet-echoes-labs512-scout-v1"
createdAt: "2026-09-04T22:21:15Z"
producer: "OpenAI Codex for MatBalez; creator adoption explicitly authorized"
promptVersion: "1.1.0"
model: "GPT-6 (Codex; self-attested model family, deployment identifier unavailable)"
disposition: "viable"
sources:
  -
    url: "https://arxiv.org/abs/2409.07222v1"
    title: "Dual-Step Optimization for Binary Sequences with High Merit Factors"
    evidence: "our approach has successfully identified new best-known binary sequences for all lengths ranging from 450 to 527"
    location: "Abstract, final paragraph; historical claim from the 2024 paper, not a claim about the present frontier"
    accessedAt: "2026-09-04"
    identifier: "arXiv:2409.07222v1"
    publicationDate: "2024-09-11"
  -
    url: "https://arxiv.org/html/2409.07222v1"
    title: "Dual-Step Optimization: exact length 512 reference"
    evidence: "512 7.0942 7.2834"
    location: "Table 2, length 512 row: length, previous merit factor, new reported merit factor. The full row is reproduced in literature/reference.json; our integer recomputation gives E = 17996. That energy is a computation, not a quoted paper value."
    accessedAt: "2026-09-04"
    identifier: "arXiv:2409.07222v1"
  -
    url: "https://arxiv.org/abs/2607.09688v1"
    title: "Prioritizing Search Space Regions in the Low Autocorrelation Binary Sequences Problem"
    evidence: "improves the previously best-known results for 35 sequence lengths"
    location: "Abstract, experimental-results paragraph. Full-text Table 3 was separately inspected; length 512 is absent from that improvement table, a bounded comparison rather than proof of a current world record."
    accessedAt: "2026-09-04"
    identifier: "arXiv:2607.09688v1"
    publicationDate: "2026-06-17"
uncertainties:
  - "Actual hosted scientific review, runtime vulnerability-policy evaluation and machine preflight must complete before publication. The patched image digest is known; no passing hosted receipt is claimed."
  - "MatBalez authorized Codex to adopt this first challenge, its disclosed thresholds and data-license obligations. This is an accountable operator adoption, not an independent human scientific or security review."
  - "The baseline is the best exact sequence found in the inspected primary sources, not an exhaustively certified current world record."
  - "Initial platform-seeded solver candidates have separate provenance. A valid artifact may score worse than the reference; no improvement is inferred from validity."
  - "Independent external scientific/security review has not occurred. Platform verification on one host and independent physical-host replication are distinct assurance states."
rejectedAlternatives:
  - "Equal-circle packing: attractive geometry and recent optimization literature, but rational-coordinate certification and record-data calibration would add complexity to a first draft."
  - "Golomb rulers: exact simple artifacts, but order 28 was proven optimal in 2022; higher-order frontier advances are poorly calibrated for this initial showcase."
repositoryPlan:
  - "checker.py: bounded integer-only artifact checker"
  - "suite/contract.json: public frozen mathematical contract"
  - "fixtures/: attributed published baseline plus valid/invalid/malformed/boundary fixtures"
  - "tests/: independent arithmetic implementation and adversarial tests"
  - "tools/reproduce.py: exact baseline reproduction, no solving"
  - "visualize.html: local data-driven waveform and autocorrelation explorer"
  - "challenge-brief.md and harness-plan.md: scientific meaning, provenance and verification"
  - "requirements.lock: no third-party dependencies"
  - "LICENSE and THIRD_PARTY_NOTICES.md: MIT first-party code; CC BY 4.0 baseline data exception"
manifest:
  verificationPolicy: "platform"
  apiVersion: "science-ladder/v1"
  kind: "ChallengeManifest"
  id: "quiet-echoes-labs512-v1"
  createdAt: "2026-09-04T22:21:15Z"
  producer: "MatBalez via Codex; platform-seeded creator adoption under explicit user authorization"
  slug: "quiet-echoes-labs512"
  title: "Quiet Echoes: the 512-bit signal challenge"
  summary: "Lower exact sidelobe energy of a 512-chip binary pulse below a reproduced literature reference. A small artifact, a vast search space, and a score anyone can recompute."
  scientificQuestion: "Can a length 512 sequence of +1 and -1 achieve aperiodic sidelobe energy E below 17996, the best inspected and reproduced published reference?"
  evidence:
    -
      url: "https://arxiv.org/abs/2409.07222v1"
      title: "Dual-Step Optimization for Binary Sequences with High Merit Factors"
      evidence: "our approach has successfully identified new best-known binary sequences for all lengths ranging from 450 to 527"
      location: "Abstract, final paragraph; historical claim from the 2024 paper, not a claim about the present frontier"
      accessedAt: "2026-09-04"
      identifier: "arXiv:2409.07222v1"
      publicationDate: "2024-09-11"
    -
      url: "https://arxiv.org/html/2409.07222v1"
      title: "Dual-Step Optimization: exact length 512 reference"
      evidence: "512 7.0942 7.2834"
      location: "Table 2, length 512 row: length, previous merit factor, new reported merit factor. The full row is reproduced in literature/reference.json; our integer recomputation gives E = 17996. That energy is a computation, not a quoted paper value."
      accessedAt: "2026-09-04"
      identifier: "arXiv:2409.07222v1"
    -
      url: "https://arxiv.org/abs/2607.09688v1"
      title: "Prioritizing Search Space Regions in the Low Autocorrelation Binary Sequences Problem"
      evidence: "improves the previously best-known results for 35 sequence lengths"
      location: "Abstract, experimental-results paragraph. Full-text Table 3 was separately inspected; length 512 is absent from that improvement table, a bounded comparison rather than proof of a current world record."
      accessedAt: "2026-09-04"
      identifier: "arXiv:2607.09688v1"
      publicationDate: "2026-06-17"
  impact: "Low-autocorrelation sequence design connects discrete optimization, Ising spin models and pulse compression. This finite benchmark measures total idealized sidelobe energy; it does not establish real radar, sonar, GNSS or quantum-system performance."
  limitations:
    - "The baseline is a 2024 preprint result published in 2025; later 2026 work was inspected. The literature search is bounded, so beating this reference is not automatically a world-record claim."
    - "One binary sequence at one length does not prove global optimality or an asymptotic merit-factor conjecture."
    - "Total sidelobe energy and peak sidelobe level are different objectives. A better score can have a higher individual peak."
    - "No Doppler, noise, bandwidth, cross-correlation, hardware or multi-target robustness is scored."
    - "The runtime is pinned to a real patched image. Creator adoption is authorized; actual hosted scientific review, vulnerability-policy checks, machine preflight and publication remain separate requirements. Local test reports do not satisfy hosted verification or independent replication."
  safetyClassification: "low-risk-computational"
  economicMode: "none"
  metric:
    name: "Aperiodic sidelobe energy"
    direction: "minimize"
    unit: "sum of squared correlations"
    quantum: "1"
    baselineTicks: "17996"
    minimumDeltaTicks: "4"
    toleranceTicks: "0"
    domainMinTicks: "256"
    domainMaxTicks: "44608256"
  hardGates:
    - "one_canonical_sequence"
    - "binary_length_512"
  milestones:
    -
      id: "first-lower-energy"
      title: "One real step beyond the reference"
      thresholdTicks: "17992"
      rationale: "At length 512 every attainable energy is a multiple of 4. This is the next possible strict improvement; novelty still requires a current literature check."
    -
      id: "one-percent"
      title: "1% less sidelobe energy"
      thresholdTicks: "17816"
      rationale: "At least 1% below the reproduced reference; integer threshold rounded down to an attainable energy residue. This is a research target, not a promised achievable score."
    -
      id: "two-point-five-percent"
      title: "2.5% less sidelobe energy"
      thresholdTicks: "17544"
      rationale: "At least 2.5% energy reduction relative to 17996; does not imply a 2.5% real receiver improvement."
    -
      id: "five-percent"
      title: "5% less sidelobe energy"
      thresholdTicks: "17096"
      rationale: "An ambitious finite-sequence target, at least 5% lower total sidelobe energy; feasibility is not established."
  deadline: "2027-03-04T23:59:59Z"
  submission:
    allowedPaths:
      - "sequence.txt"
    allowedExtensions:
      - ".txt"
    maxBytes: 513
    maxFiles: 1
    license: "CC-BY-4.0"
  validator:
    profile: "artifact-checker-v1"
    entrypoint:
      - "/usr/local/bin/python3"
      - "/sl/challenge/checker.py"
    dependencyLock: "requirements.lock"
    runtimeImageDigest: "sha256:a8136bf6f5082a72776f2565f279a6c336b462da483aa54cfa81586ad20705fa"
  suite:
    visibility: "public"
    path: "suite"
  resources:
    class: "cpu-small"
    vCpu: 1
    memoryMb: 128
    timeoutSeconds: 5
    maxOutputBytes: 4096
  fixtures:
    -
      name: "baseline"
      path: "fixtures/baseline"
      expectedOutcome: "valid"
      expectedTicks: "17996"
    -
      name: "valid"
      path: "fixtures/valid"
      expectedOutcome: "valid"
      expectedTicks: "43776"
    -
      name: "invalid"
      path: "fixtures/invalid"
      expectedOutcome: "hard_gate_failed"
    -
      name: "malformed"
      path: "fixtures/malformed"
      expectedOutcome: "invalid_output"
    -
      name: "empty"
      path: "fixtures/empty"
      expectedOutcome: "hard_gate_failed"
    -
      name: "oversized"
      path: "fixtures/oversized"
      expectedOutcome: "invalid_output"
    -
      name: "maximum"
      path: "fixtures/maximum"
      expectedOutcome: "valid"
      expectedTicks: "44608256"