{"id":"cbcdc958-77e2-4342-b87c-a93ee8448443","arxiv_id":"2606.09593","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"PHDAG achieves depth-independent O(1) gas cost for appends in on-chain provenance registries, outperforming incremental Merkle trees beyond very small depths, with linear-time trustless reconstruction from event logs.","lead":"The paper defines the Parent-Hash DAG (PHDAG) structure for append-only provenance trees on blockchains and shows that each append costs a fixed amount of gas regardless of how many entries already exist. A smart generalist might read it to see whether constant-cost on-chain registries are feasible for tracking artifact history at scale.","discovery_kind":"unclear","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"The reader's weakest_assumption correctly isolates the implementation precondition, yet the empirical data already falsifies the practical risk of hidden scaling costs within the reported range. Because the full manuscript is stipulated as available and the abstract plus results contain no internal contradictions or unstated dependencies that would invalidate the O(1) bound, the precondition does not constitute a load-bearing gap. The reader's UNVERDICTED status is attributable to abstract-only review rather than any flaw in the argument itself.","tokens_in":1767,"tokens_out":298,"duration_ms":24302,"concrete_test":"Re-execute the append gas measurements on Base Sepolia for depths 1-50 using the same contract bytecode; if any measurement deviates by more than 100 gas from the reported mean, the invariance claim requires revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim rests on formalizing PHDAG append as strictly O(1) gas via a fixed number of writes exclusively to untouched storage slots. The provided abstract states this pattern explicitly, derives a stochastic model for the IMT comparator with closed-form mean/variance, and reports empirical invariance at 76,276 gas (std ~6) across depths 1-25 on Base Sepolia. These elements are internally consistent; the observed constancy directly corroborates the absence of depth- or size-dependent overheads in the tested regime.","agreement_with_reader":"disagree"},"referee_report":{"model":"grok-4.3","summary":"The paper claims to introduce the Parent-Hash DAG (PHDAG) as a data structure for provenance trees on blockchain, where appends are performed with a constant number of storage writes to untouched slots, leading to O(1) gas cost independent of size and depth. It develops a stochastic model for the incremental Merkle tree (IMT) with closed-form mean and variance, validates both with empirical measurements on Base Sepolia showing constant cost for PHDAG at 76,276 gas and linear growth for IMT, and claims linear-time trustless reconstruction from event logs.","tokens_in":1864,"tokens_out":292,"duration_ms":25619,"significance":"If the results hold, the work is significant as it provides a constant-cost alternative to IMT for on-chain append-only registries, with the empirical invariance and closed-form expressions for the comparator being notable strengths. This could enable more scalable provenance infrastructure on public blockchains. The reproducible nature of the empirical runs on a public testnet adds to the reliability of the findings.","major_comments":[],"minor_comments":[{"comment":"The standard deviation is given as 'about 6 gas'; providing the exact value and the number of trials would improve precision.","section":null},{"comment":"Ensure that the closed-form expressions for the IMT mean and variance are presented clearly with all variables defined.","section":null}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the positive assessment of the PHDAG paper, the accurate summary of its contributions, and the recommendation for minor revision. No major comments were provided in the report.","responses":[],"tokens_in":1262,"tokens_out":57,"duration_ms":8136,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that this paper isolates the parent-hash DAG as a way to append to an on-chain registry at fixed gas cost, measured at 76,276 gas with a standard deviation of about 6 across depths 1-25 on Base Sepolia. The incremental Merkle tree comparator scales linearly, and they supply closed-form mean and variance for its cost.\n\nWhat stands out is treating the PHDAG pattern as its own primitive, which the abstract says had not been done before, then running the first head-to-head test at relevant depths and deriving the stochastic model for IMT. The reconstruction from event logs in linear time without off-chain state is a practical addition. The measurements line up with the constant-cost claim and the crossover point sits well below the production registries they surveyed.\n\nThe soft spots are limited. The abstract omits the explicit gas-cost equations and contract-level write details, so the constant number of untouched-slot writes rests on the reported invariance rather than a visible derivation. That said, the empirical data directly supports the absence of depth-dependent overhead in the tested range, and the circularity burden stays low because the runs are fresh. No load-bearing assumption appears contradicted by the numbers.\n\nThis is for people working on provenance systems, on-chain registries, or other append-only blockchain structures. A reader in distributed systems or applied cryptography gets usable constants and a clear model to compare against.\n\nSend it to referees. The evidence is concrete enough to evaluate and the central claim holds up on the data provided.","headline":"PHDAG gives a flat 76k-gas append for on-chain registries while IMT grows linearly, with empirical runs and closed forms backing the comparison.","tokens_in":2369,"tokens_out":386,"would_cite":true,"duration_ms":16658,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Parent-hash DAGs allow appends to on-chain registries at fixed gas cost independent of size or depth.","keywords":["parent-hash DAG","PHDAG","on-chain registry","gas cost","constant-time append","incremental Merkle tree","provenance tree","blockchain storage"],"falsifier":"Deploy the PHDAG contract on an EVM chain and measure gas usage for successive appends; if measured costs rise materially with depth or total entries, the constant-time claim does not hold.","tokens_in":2655,"feed_emoji":"🔗","tokens_out":769,"duration_ms":18156,"temperature":0.7,"pith_summary":"The paper establishes that a parent-hash directed acyclic graph structure lets each new entry into an on-chain provenance registry perform only a constant number of storage writes to untouched slots. This produces gas costs that stay flat regardless of how many prior entries exist or how deep the structure grows. A sympathetic reader would care because it removes the scaling penalty that forces registry operators to choose between high fees and small size. The authors contrast this with incremental Merkle trees, for which they derive a stochastic cost model showing linear growth in expected gas with depth, and they confirm the difference with measurements across depths 1 to 25 on Base Sepolia. They also show that the full registry state can be rebuilt trustlessly from public logs in time linear in the number of entries.","feed_headline":"PHDAG appends cost fixed gas independent of registry depth","feed_subtitle":"Constant 76k gas per insert on Base Sepolia; IMT costs rise linearly and cross over at low depths","key_machinery":"The parent-hash directed acyclic graph (PHDAG) pattern, in which each append writes a constant number of new hashes to previously untouched storage slots.","core_discovery":"PHDAG append is formalized as O(1) in gas cost because each insertion writes a fixed number of new parent hashes exclusively into previously untouched storage slots; the cost is therefore independent of registry size and tree depth. Empirical runs on Base Sepolia show PHDAG costs remain at 76,276 gas with standard deviation near 6 gas, while IMT per-insert cost grows linearly with depth. Closed-form expressions are derived for the mean and variance of IMT cost as a random variable over leaf index. The point at which IMT becomes more expensive lies below the depths of all surveyed production registries. Registry state can be reconstructed from event logs in linear time with no off-chain data.","pith_inferences":["The same constant-write pattern could be applied to other append-only structures that currently rely on Merkle trees for on-chain verification.","If gas invariance holds across EVM-compatible chains, operators could migrate registries without re-evaluating per-operation economics.","Linear-time reconstruction from logs remains feasible only if the total number of entries stays within practical query budgets for the target chain."],"forward_implications":["On-chain registries using PHDAG can grow arbitrarily large while keeping per-append gas cost fixed.","The cost advantage over incremental Merkle trees appears at depths already exceeded by every production registry examined.","Trustless reconstruction of the entire registry from public event logs succeeds in time linear in the number of entries.","The stochastic model supplies explicit mean and variance formulas for the gas cost of each IMT insertion."],"fun_headline_variants":["PHDAG locks gas at 76276 constant across all depths","Fixed 76k gas for PHDAG appends beats IMT linear rise","PHDAG proves O(1) gas cost independent of tree depth","76k gas invariant for PHDAG; IMT grows linearly with depth","PHDAG append cost stays fixed on Base Sepolia tests"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The implementation can be realized so that every append performs only a constant number of storage writes exclusively to previously untouched slots, without hidden state-dependent overheads or additional writes that scale with current depth or size.","fun_headline_variants_meta":{"raw":{"variants":["PHDAG locks gas at 76276 constant across all depths","Fixed 76k gas for PHDAG appends beats IMT linear rise","PHDAG proves O(1) gas cost independent of tree depth","76k gas invariant for PHDAG; IMT grows linearly with depth","PHDAG append cost stays fixed on Base Sepolia tests"]},"model":"grok-4.3","cost_usd":0.003719,"raw_usage":{"total_tokens":1964,"prompt_tokens":738,"num_sources_used":0,"completion_tokens":91,"cost_in_usd_ticks":37187000,"prompt_tokens_details":{"text_tokens":738,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1135,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":738,"tokens_out":91,"duration_ms":6447,"temperature":1.0,"reasoning_tokens":1135,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T14:51:28.443454+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Deploy the PHDAG contract on an EVM chain and measure gas usage for successive appends; if measured costs rise materially with depth or total entries, the constant-time claim does not hold.","supporting_citations":[],"review_version":1}