aere-research/results/kat-results-challenger.json
Aere Network 6cb0140fae Republished from a clean root: the compiled artifact is gone from history, and the local line of work joins the sanitized public line
The public history carried kat/__pycache__/mlkem768_reference.cpython-314.pyc,
a compiled Python artifact embedding the operator's absolute local path. Text
secret scanners do not read compiled binaries, which is exactly how it slipped
through, and removing it from the tip would have left it reachable through the
old root commits. So this repository is republished from a single clean root.

This root also carries, from the previously unpublished line of work:
- corrected LICENSE year, LICENSING.md, VERIFY-POLICY.md, and
  CITATIONS-UNRESOLVED.md remeasured 2026-08-11 (101 paths, README aligned)
- O-018: run_consensus_verification.py ran 19 of 29 models and reported PASS;
  it now runs all 29, and computemarket_smt.py gains resolveByTimeout /
  reclaimUnsettled cases plus a negative control
- O-006: the word 'audited' removed from next to Bouncy Castle, twice, after a
  concurrent edit resurrected it
- O-014: prior art named and dated - Algorand's native falcon_verify shipped
  about ten months before AERE's precompiles; the primacy claim is withdrawn
  where it was implied
- bench/ scripts parametrized so they actually run for an outsider (the
  earlier textual sanitization left $STAGING unexpanded inside Python strings)
- AIP-2/AIP-3 errata with measured figures, spec remeasurements at 2026-08-01,
  and the spec-zk-stack retractions (owner is an operational key, not the
  Foundation; 'maximally sound' withdrawn; aggregator V1 deprecated)
The redacted bench-host environment files from the sanitized line are kept
exactly as published; the unredacted local variants are not carried.
2026-08-15 13:52:14 +03:00

19 lines
3.2 KiB
JSON

{
"component": "Fiat-Shamir transcript / Poseidon2 DuplexChallenger (port spec component (f))",
"precompile": "0x0000000000000000000000000000000000000ae8",
"scope": "port spec section 7; the duplex-sponge challenger DuplexChallenger<BabyBear, Perm, WIDTH=16, RATE=8> and the GrindingChallenger check_witness: observe (overwrite-mode absorb), sample (squeeze, pop from the end of the rate buffer), sample_bits (low-bits query-index reduction), check_witness (observe witness then sample_bits==0). Wired behind the still-fail-closed top level.",
"validates": "CONFORMANCE: ground truth emitted by executing the pinned Plonky3 p3-challenger 0.4.3-succinct DuplexChallenger + GrindingChallenger (and the p3-fri verify_shape_and_sample_challenges transcript ORDER). (1) a fully-scripted observe/sample transcript reproduces every sampled base + F_{p^4} ext value, the full 16-lane sponge state, and a check_witness accept/reject table byte-for-byte; (2) the FRI betas + query indices + grinding acceptance are DERIVED from the transcript (commit_phase_commits + final_poly + pow_witness) and match the pinned p3-fri challenger AND the confirmed FRI ground truth; (3) END TO END: those derived betas/indices, fed into the CONFIRMED FRI verify_query (component (d)), accept the real FRI proof and reject a tampered beta. Python/Node/standalone-Java are byte-identical.",
"does_not_validate": "the SP1 recursion-AIR constraint evaluation (component (e)): the reduced openings, DEEP zeta/alpha challenges, quotient consistency, and vk binding. That is un-ported (StarkConstraints.evaluateAtZeta = UNAVAILABLE). A real exported SP1 v6.1.0 proof's transcript (vkey digest, trace/quotient commitments, opened values order) is still [MEASURE].",
"top_level_verifier": "unchanged, still FAIL-CLOSED (returns EMPTY for every input). Even with component (f) confirmed and Challenger.spongePorted flipped true, StarkConstraints.evaluateAtZeta returns UNAVAILABLE (component (e) un-ported), so verify() returns UNAVAILABLE before any query loop and computePrecompile returns EMPTY. ACCEPT is unreachable. DO NOT ACTIVATE.",
"confirmed_config": "DuplexChallenger<BabyBear, Poseidon2<BabyBear,...,16,7>, WIDTH=16, RATE=8>; Witness=BabyBear; Challenge=BinomialExtensionField<BabyBear,4>; sample pops from the END of the rate buffer (Vec::pop); sample_bits masks the LOW `bits` of as_canonical_u64; check_witness = observe(witness) then sample_bits(bits)==0.",
"pinned_conformance_target": "p3-challenger 0.4.3-succinct (duplex_challenger.rs, grinding_challenger.rs), transcript order p3-fri 0.4.3-succinct verifier.rs (verify_shape_and_sample_challenges), sponge = confirmed Poseidon2-BabyBear (component (b)); byte-identical to the repo Cargo.lock. Extractor: pq-stark/challenger-extractor -> pq-stark/challenger_ground_truth.json.",
"cross_language": "Python (challenger_reference.py), Node (challenger_reference.mjs), standalone Java (ChallengerSelfTest.java) emit byte-identical shared vectors.",
"flags": {
"Challenger.spongePorted": true,
"note": "flipped true: component (f) confirmed. StarkConstraints.evaluateAtZeta stays UNAVAILABLE (component (e)), so the top level stays fail-closed and ACCEPT is unreachable."
},
"notes": [],
"passed": 44,
"failed": 0,
"total": 44
}