{"id":"80027a8e-131a-49f4-88b9-d218992fdbef","arxiv_id":"1908.06921","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"The paper proposes a two-phase blockchain trust model for 3D design attestation, with payments derived from an existing peer prediction result.","lead":"This paper designs a blockchain marketplace where 3D design files are tested by paid community reviewers and get an integrity score before being sold. It combines a reputation system with reward and penalty payments to encourage truthful reviews, and describes how the process runs on Ethereum smart contracts.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (6)-(7) define τ_r and τ_p by two different formulas that are not algebraically equal, leaving the claimed incentive-compatible payments undefined and the central Nash-equilibrium guarantee unsupported.","rationale":"The reader's weakest assumption identifies both the unproven reduction and the algebraic inconsistency in Eqs. (6)-(7). The algebraic inconsistency is the more concrete and decisive defect: the paper prints two different formulas for the same payment parameters, and they are not equal for any q* in (1/2, 1). This is not a matter of external disagreement; it is an internal contradiction in the statement of the central mechanism. Since the claimed Nash-equilibrium guarantee depends on choosing the right payments, and the paper does not specify which formula is authoritative, the guarantee is not established as stated. The unproven reduction to Witkowski et al. compounds the problem, because even with a consistent payment formula there is no demonstrated equivalence between the pairwise comparison game and the weighted-majority voting game. Both defects are present in Section V-D. The paper does provide a concrete blockchain architecture and a commit-reveal protocol, but these do not repair the missing incentive proof. The self-acknowledged limitations regarding coalitions and unverifiable effort further weaken the practical claim, but the central theoretical claim is already unsupported by the payment-formula contradiction. Therefore, the reader's REJECT verdict remains appropriate.","tokens_in":12053,"tokens_out":3352,"duration_ms":33563,"concrete_test":"Set q* = 0.75 and C* = 1. Compute x* = 2q* - 1 = 0.5. Evaluate the two expressions in Eq. (6): 2C*/((x*)^2 + x*) = 2/0.75 ≈ 2.667, and C*/(2(q*)^2) = 1/(2·0.5625) ≈ 0.889. Since these differ, the paper's own payment formula is ambiguous. Then, taking each candidate value separately, re-derive the equilibria of the weighted-majority game defined by Eqs. (4)-(5), including the option to abstain, and check whether truthful effort is a Nash equilibrium under either payment. If neither candidate satisfies the conditions of Witkowski et al.'s theorem adapted to this game, the central incentive guarantee fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that with payments τ_r and τ_p chosen as in Section V-D, rational, selfish, and independent agents will invest effort and answer truthfully, making truthfulness a Nash equilibrium. The paper imports this result from Witkowski et al. [11] by 'logically reducing' their pairwise peer-prediction game to the weighted-majority game here. Two load-bearing defects undermine this. First, the reduction is asserted, not proved; the paper changes the comparison from a random peer match to a comparison against the weighted majority of all other players, and no argument shows that the equilibrium properties of [11] survive this change. Second, the payment formulas themselves are internally inconsistent. Eq. (6) simultaneously states τ_r = 2C*/((x*)^2 + x*) and τ_r = C*/(2(q*)^2), with x* = 2q* - 1. For general q*, these are different: for q* = 0.75, the first gives 2C*/0.375 ≈ 2.67C*, while the second gives C*/(2·0.5625) ≈ 0.89C*. Eq. (7) has the same structure for τ_p. Since the equilibrium theorem in [11] is parameterized by specific payment values, an ambiguous or false identification means the paper has not actually connected the borrowed theorem to the formulas it prints. Thus the stated values of τ_r and τ_p do not, as written, constitute a well-defined incentive scheme, and the central claim that truthfulness is a Nash equilibrium is not established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a blockchain-based decentralized marketplace for the attestation of 3D-printing design files. The core mechanism is a two-phase trust model: an evaluation phase and a feedback phase, in which players vote on the validity of a design, a reputation-weighted majority score is computed, and players are rewarded or penalized according to whether their vote agrees with the majority outcome. The authors claim that, with suitable reward and penalty values, rational, selfish, and independent agents are incentivized to invest effort and report truthfully, making truthfulness a Nash equilibrium. The claimed incentive-compatible payments are taken from a peer-prediction model by Witkowski et al. and are presented in Eqs. (6) and (7). The paper also describes an Ethereum smart-contract implementation with a commit-reveal voting scheme and a semi-trusted manager for fair exchange of designs for evaluations.","tokens_in":12359,"tokens_out":6351,"duration_ms":66744,"significance":"If the central game-theoretic claim were rigorously established, the paper would make a useful contribution: it applies peer-prediction ideas to a concrete, high-stakes crowdsourced validation problem, and it sketches a blockchain implementation with public verifiability. The paper has several praiseworthy features: it builds on a well-known incentive-compatibility result rather than inventing an ad hoc mechanism, it uses a commit-reveal scheme to reduce herding, and it clearly describes the protocol flow and smart-contract logic. However, the central equilibrium claim is not supported as written. The payment formulas in Eqs. (6) and (7) are algebraically inconsistent, so the reward and penalty values are undefined. Moreover, the claimed reduction of Witkowski et al.'s pairwise peer-prediction game to this paper's aggregate weighted-majority comparison is asserted but not proved. Because the paper's main contribution is precisely this trust model, these issues are load-bearing and currently prevent acceptance.","major_comments":[{"comment":"The two formulas printed for the reward τ_r are not equal except at q*=0, which is outside the domain q*>1/2. Substituting x* = 2q* - 1 gives 2C*/(x*^2 + x*) = C*/(q*(2q* - 1)), whereas the second expression is C*/(2(q*)^2); for q* = 0.75, these evaluate to approximately 2.67C* and 0.89C*, respectively. The same inconsistency appears in Eq. (7) for τ_p. Since the equilibrium theorem borrowed from Witkowski et al. is parameterized by specific payment values, the paper does not actually define an incentive scheme, and the claim that the displayed payments induce truthfulness as a Nash equilibrium is unsupported.","section":"Section V-D, Eqs. (6)-(7)"},{"comment":"The reduction from Witkowski et al.'s model to the model of this paper is asserted, not proved. The original model compares an agent's answer to the answer of a randomly selected peer, whereas this paper compares the agent's weighted vote with the weighted majority of all other players. In the aggregate comparison, the reference is endogenous: a player's own reputation weight and vote can affect the comparison, and the distribution of the reference signal is different from the pairwise setting. The sentence 'we reduce the complexity by assuming that the player's vote is compared to another peer with the weighted answer' does not establish that the equilibrium properties of [11] carry over. This gap affects the central claim of the paper, namely that the proposed reward/penalty system makes truthfulness a Nash equilibrium.","section":"Section V-D"},{"comment":"The identification of the effort cost C* with the cost of the material needed to 3D print the design is not justified. In the application, effort includes printing time, testing or integration of the printed object, and the evaluator's expertise, not merely material cost. If C* underestimates the true effort cost, the computed payments τ_r and τ_p will be too low to satisfy the incentive-compatibility thresholds from Witkowski et al., so the claimed equilibrium would not hold. The authors need either to justify this assumption with a cost model or to treat C* as a parameter that includes all costs of effort.","section":"Section V-D"}],"minor_comments":[{"comment":"The definition of τ_p as a negative value in Eq. (7) conflicts with the text stating that players 'will pay a penalty of τp' and with Algorithm 1's requirement that a player's collateral be 'at least τp'; if τ_p is negative, this condition is vacuous. Please clarify by using |τ_p| or by defining the penalty as a positive amount.","section":"Section V-C and VI-B"},{"comment":"The condition 'discard if playerStates[j].received ≠ TRUE' appears to be a typo: it should refer to playerStates[pi].received[j], since the contract otherwise stores per-player, per-design received flags.","section":"Algorithm 1, commit handler"},{"comment":"The final-score formula in Eq. (4) uses w_i(j) = |T(i)|/Σ_k |T(k)|, but Algorithm 2 does not show how the contract computes |T(i)| or updates the stored weight according to Eq. (3). The pseudocode should be aligned with the formulas it claims to implement.","section":"Section V-B and Algorithm 2"},{"comment":"The outcome r_j = 0, in which the game is annulled because FS(j) falls in [1-q*, q*], is not handled in the compensation rules; the paper should state whether players receive rewards, penalties, or nothing in this case.","section":"Section V-B"}],"recommendation":"reject","confidential_remarks":"This manuscript has a promising application and a plausible architecture, but the central theoretical contribution is not established: the printed payment formulas are self-contradictory, and the key reduction to Witkowski et al. is not proved. These are not merely presentation issues; they concern the main claimed result. A revision would require substantial new analysis (a formal equivalence proof or a corrected mechanism), not just local edits, so I recommend rejection. The authors might resubmit if they can rigorously connect their voting rule to an appropriate incentive-compatibility theorem and provide consistent payment formulas."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper is a good example of a system idea that outruns its theory. The problem is real—3D design files need distributed validation—and the proposed architecture (two-phase evaluation/feedback, reputation-weighted majority voting, commit-reveal on Ethereum) is a sensible starting point. The smart-contract pseudocode is concrete, and the authors are honest about what they don't model: coalitions, proof-of-effort, and reputation-concentration attacks.\n\nBut the central incentive guarantee does not hold as written. Eq. (6) defines τ_r two ways that are not equal: τ_r = 2C*/(x*^2+x*) and τ_r = C*/(2q*^2), with x*=2q*-1. For q*=0.75, the first gives 2.67C*, the second 0.89C*. Eq. (7) has the same problem for τ_p. So the payment values that are supposed to induce truthfulness are undefined. The paper says it \"logically reduces\" Witkowski et al.'s model to this one, but the comparison is changed from a random peer to a weighted majority of all other players, and no proof is given that the equilibrium properties survive that change. The theorem is parameterized by specific payments; since those payments are incoherent, the theorem was never actually connected to the system.\n\nThere is also the questionable assumption that effort cost equals material cost. Validating a 3D design involves time, expertise, and possibly destructive testing—not just filament. And despite the implementation language, there is no running code, no simulation, no gas-cost analysis, and no adversarial evaluation. Those are legitimate gap but secondary to the algebraic failure.\n\nWho benefits? Anyone working on decentralized attestation or blockchain-based peer prediction will find the architecture thought-provoking, and the mistakes are instructive. But as a contribution claiming a Nash-equilibrium guarantee, it fails. It deserves peer review because the problem is significant and the system design has merit, but the incentive section needs to be rebuilt—or honestly reframed as a conjecture.","headline":"A well-motivated but mathematically broken attestation marketplace: the payment formulas contradict each other and the equilibrium reduction is unproved.","tokens_in":12877,"tokens_out":2849,"would_cite":false,"duration_ms":30480,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A reward-and-penalty scheme tied to material cost and a quality threshold makes truthful validation of 3D designs a Nash equilibrium in a blockchain marketplace.","keywords":["additive manufacturing","3D printing","blockchain","smart contracts","truthfulness","peer prediction","weighted majority voting","reputation systems"],"falsifier":"Compute both expressions the paper gives for the reward in Eq. (6) at a concrete threshold such as $q^*=3/4$ (so $x^*=1/2$): the first reads $2C^*/(1/4+1/2)=8C^*/3$ and the second reads $C^*/(2\\cdot(9/16))=8C^*/9$, which are not equal; a reader could therefore settle the mechanism's definition by determining which payment is actually used and then simulating the two-phase game to test whether truth-telling remains a Nash equilibrium.","tokens_in":11844,"feed_emoji":"🖨️","tokens_out":7574,"duration_ms":74531,"temperature":0.7,"pith_summary":"The paper argues that a distributed marketplace can reliably validate 3D-printable designs without a central authority by paying rational, selfish agents to test designs and tell the truth. It proposes a two-phase process: an evaluation phase in which testers vote on a design, and a feedback phase from buyers that acts as a safety net, with reputation-weighted majority voting deciding whether a design is valid. The central claim is that a carefully chosen reward/penalty system, with payments derived from the cost of printing material and a quality threshold, makes truthful effort a Nash equilibrium. If that is right, buyers could trust community-vetted 3D designs, and the same incentive mechanism could be reused for other expert-elicitation marketplaces such as open-source code review.","feed_headline":"Truthful 3D reviews become a Nash equilibrium with the right payments","feed_subtitle":"If right, buyers can trust community-validated 3D designs without any central authority.","key_machinery":"The load-bearing mechanism is the payment pair $(\\tau_r, \\tau_p)$ together with the reputation-weighted final score. The final score is $FS(j) = \\tfrac{1}{2}\\bigl(\\sum_i a_i(j)\\mathrm{rep}(i)w_i(j) \\big/ \\sum_i \\mathrm{rep}(i)w_i(j) + 1\\bigr)$, and the design's verdict $r_j$ is decided by comparing $FS(j)$ with the threshold $q^*$. A player is rewarded when the sign of their weighted vote matches the sign of the weighted vote of the rest of the players, and penalized otherwise; reputation is updated from agreement with past weighted majorities. The argument's work is done by importing an equilibrium theorem from a cited peer-prediction game and applying it, by asserted logical reduction, to this more complex weighted-majority voting game.","core_discovery":"The paper claims that by setting evaluation rewards and penalties to specific values, and by comparing each player's weighted vote to the weighted majority of the remaining players, a marketplace of rational, selfish, and independent agents will choose to invest effort and report truthfully. The reward is given as $\\tau_r = 2C^*/(x^{*2}+x^*)$ and the penalty as $\\tau_p = -2C^*/(x^{*2}+x^* - \\varepsilon)$, where $C^*$ is the cost of printing material and $q^* = (x^*+1)/2$ is the quality threshold. The paper identifies these expressions with $C^*/(2q^{*2})$ and $-C^*/(2q^{*2}-\\varepsilon)$, asserting that this choice imports a previously proven peer-prediction equilibrium result into the proposed weighted-majority setting. It further claims that the whole process, including reputation calculation, final-score computation, and payments, can be run transparently by a smart contract on a blockchain.","pith_inferences":["If the equality between the two payment formulas in Eqs. (6)-(7) is not restored, a real deployment would have to choose one schedule; the two choices pay very different amounts for the same threshold, so the equilibrium argument may not transfer to either.","The paper's own discussion concedes that coalitions of players are outside the model; a natural testable extension is a vote-correlation detector that flags groups who always vote together and reduces their influence, as the authors themselves identify as future work.","A proof-of-print mechanism, in which a trusted component of the 3D printer attests that a design was physically printed, would make the cost-of-effort assumption more concrete and could be used to scale each player's vote weight.","The generic incentive design suggests an immediate experiment: run the two-phase voting game in a simulated marketplace with artificial agents and check whether the specified payments actually induce effort and truthful reporting before any blockchain deployment."],"forward_implications":["If the equilibrium claim holds, a vendor can post a design for free community testing, pay testers only when their verdict matches the weighted majority, and sell only designs that pass both the evaluation and feedback phases.","Because rewards scale with the cost of printing material $C^*$, expensive or resource-intensive designs would carry larger rewards, matching compensation to the effort required for physical attestation.","The commit-reveal voting scheme prevents late voters from copying earlier votes, and the collateral requirement makes abstention or deliberate non-revelation costly.","The same incentive core could be packaged as a generic attestation service for other tasks without known ground truth, such as community verification of open-source code or bug reports.","Publicly verifiable reputation and final scores would let buyers compare designs by an integrity score that no single authority can silently alter."],"supporting_citations":[{"why":"Supplies the equilibrium theorem that suitable reward and penalty amounts elicit effort and truthful reporting, which the paper adopts for its payment values.","marker":"[11]"},{"why":"Provides the earlier output-agreement model with agent prescreening that the imported equilibrium result builds on and relaxes.","marker":"[10]"},{"why":"Gives the reputation calculation that the paper adapts into Eq. (2) for each player's reputation.","marker":"[9]"},{"why":"Establishes trust and reputation as core e-commerce concepts that motivate the marketplace design.","marker":"[8]"},{"why":"Specifies Ethereum, the smart-contract infrastructure on which the design voting and payment logic is implemented.","marker":"[15]"}],"fun_headline_variants":["Blockchain marketplace makes 3D design reviews truthful via smart contracts","Incentive scheme ensures truthful 3D design reviews on blockchain","Reward and penalty system keeps 3D design marketplace truthful","Distributed attestation of 3D designs via blockchain incentives"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The truthfulness guarantee depends on the assumption, asserted rather than proved in Section V-D, that this weighted-majority two-phase voting game is the same game as the peer-comparison game whose equilibrium result is imported, and that the stated reward and penalty formulas are consistent with that result.","fun_headline_variants_meta":{"raw":{"variants":["Blockchain marketplace makes 3D design reviews truthful via smart contracts","Incentive scheme ensures truthful 3D design reviews on blockchain","Reward and penalty system keeps 3D design marketplace truthful","Distributed attestation of 3D designs via blockchain incentives"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000425,"raw_usage":{"total_tokens":2172,"prompt_tokens":933,"completion_tokens":1239,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":549,"completion_tokens_details":{"reasoning_tokens":1166}},"tokens_in":549,"tokens_out":1239,"duration_ms":10159,"temperature":1.0,"reasoning_tokens":1166,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:31:08.412129+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute both expressions the paper gives for the reward in Eq. (6) at a concrete threshold such as $q^*=3/4$ (so $x^*=1/2$): the first reads $2C^*/(1/4+1/2)=8C^*/3$ and the second reads $C^*/(2\\cdot(9/16))=8C^*/9$, which are not equal; a reader could therefore settle the mechanism's definition by determining which payment is actually used and then simulating the two-phase game to test whether truth-telling remains a Nash equilibrium.","supporting_citations":[{"cited_title":"Dwelling on the negative: Incentivizing effort in peer prediction,","cited_arxiv_id":null,"evidence_quote":"Supplies the equilibrium theorem that suitable reward and penalty amounts elicit effort and truthful reporting, which the paper adopts for its payment values."},{"cited_title":"Crowdsourced judgement elicitation with endogenous proficiency,","cited_arxiv_id":null,"evidence_quote":"Provides the earlier output-agreement model with agent prescreening that the imported equilibrium result builds on and relaxes."},{"cited_title":"Peertrust: Supporting reputation-based trust for peer-to-peer electronic communities,","cited_arxiv_id":null,"evidence_quote":"Gives the reputation calculation that the paper adapts into Eq. (2) for each player's reputation."},{"cited_title":"Trust and e-commerce: a study of consumer perceptions,","cited_arxiv_id":null,"evidence_quote":"Establishes trust and reputation as core e-commerce concepts that motivate the marketplace design."},{"cited_title":"A next-generation smart contract and decentralized application platform,","cited_arxiv_id":null,"evidence_quote":"Specifies Ethereum, the smart-contract infrastructure on which the design voting and payment logic is implemented."}],"review_version":1}