Provenance technology is overwhelmingly built to answer the question "is this product fake?". We argue that this framing systematically disadvantages the producers it is nominally meant to protect. Anti-counterfeiting is an arms race in which the defender must be right every time, and, more importantly, it places the cost of proof on whoever must comply-which is why audit-based certification prices out smallholders, and why due-diligence regimes such as the EU Deforestation Regulation now put their market access at risk. We invert the objective. Rather than making forgery impossible, we make honesty cheap to demonstrate and differentially expensive to imitate. We present the design of Trace, an implemented and publicly available provenance system for smallholder farmers and artisan makers, and articulate its central architectural idea: asymmetric trust infrastructure. Key management is a burden proportional to institutional capacity, so Trace places it only where that capacity exists. Origin records are keyless and content-addressed, created offline on a commodity phone by a producer who has no keys to lose. Intermediate custody-carriers, warehouses, exporters-is cryptographically signed, because logistics firms can run a root key and delegate to staff devices. The final receipt at the last mile is keyless again, because a consumer will not enrol. Both human ends of the chain stay frictionless; signatures appear only in the institutional middle. We give a per-layer threat model that states precisely what each mechanism does and does not guarantee, and we are explicit that origin authenticity is not a cryptographic property of this system but a social one, resting on witnesses and accumulated multi-party history. We argue this is the correct place for it to rest. We conclude with a detailed evaluation protocol; the studies it specifies have not yet been run, and we present this as a design paper rather than an empirical one.
Smart contract vulnerability detection usually uses public datasets for training and evaluation. However, using datasets that contain duplicate and highly similar pairs can bring implicit data leakage between the training and test sets. This may affect the reliability of the evaluation results. To address this issue, this paper constructs a binary classification task related to reentrancy vulnerabilities based on two public datasets, ScrawlD and DIVE. It proposes an overlap-aware evaluation framework for fair evaluation. The framework further identifies data overlap at two distinct tiers: duplicate samples and high-similarity pairs. Two evaluation settings, random split and strict split, are constructed. Experiments are conducted using Logistic Regression, Linear Support Vector Machine (SVM), and Multinomial Naive Bayes (MultinomialNB). Results show that sample overlap exists in both datasets, with a higher degree of overlap in DIVE. And sample leakage between the training and test sets has been eliminated effectively by a strict split. Further analysis reveals that the impact of a strict split on model performance is dataset-dependent. It changes less on ScrawlD but decreases significantly on DIVE. The findings suggest that conventional random splitting tends to inflate performance metrics when sample overlap occurs.
Maximal Extractable Value (MEV) has been a longstanding unfairness and volatility in Ethereum's final execution, as there are opportunities for transaction ordering to allow for private gains that precede the observation of ordinary users. This study builds a framework for MEV identification, adaptive defense, and short-horizon prediction based on graphs. Records of transactions from MEV labels, bundle-level observations, and Ethereum on-chain data are organized into a heterogeneous transaction graph. A Relational Graph Convolutional Network (RGCN) is employed to learn representations of accounts and transactions that are aware of their relations, and the learned representations are integrated with engineered transaction features in an eXtreme Gradient Boosting (XGBoost) classifier. The defense module applies incremental updates with contrastive self- supervision in order to deal with the evolving nature of attacks. Additionally, a Temporal Graph Neural Network estimates the near-future MEV risk based on the historical graph states. Experimental results show that the graph-based design outperforms traditional classifiers, with the highest accuracy of 91.6% and the highest F1 score of 89.8%; while the temporal modelling gives better and more stable early-warning accuracy as the prediction horizon grows, with the best accuracy of 88.7% at the 10th horizon.
We develop a Bitcoin Polar Pricing Model that transforms Bitcoin prices into polar coordinates to identify, price, and forecast cyclical dynamics. Rather than imposing the four-year halving cycle, we estimate it endogenously: three independent methods converge on 3.86 years, and Bitcoin sits closer to the 1,461-day halving benchmark than Ethereum or the S&P 500 placebos under every method. The model explains 94% of Bitcoin's log-price variation, with significant within-cycle Fourier structure. Apparent predictability rises with horizon, a pattern we interpret cautiously given known overlappingwindow biases. Collectively, the polar pricing model offers a legitimate, economically grounded framework for pricing Bitcoin.
Do financial markets have memory: is every price movement independent from previous prices? This paper investigates the extent to which Bitcoin's market efficiency has transitioned towards the random walk model (H = 0.5), in comparison to the S&P 500, measured using a sliding window Hurst Exponent analysis. Bitcoin, a new asset with relatively minimal institutional surveillance, was expected to hold significant memory, while the S&P 500 serves as a mature, efficient benchmark. Daily prices from January 5, 2017 to March 27, 2026 were processed into log returns and analyzed using Rescaled Range analysis across a 252-day sliding window, producing 2,069 H values per asset. Markov chains and steady-state vectors were then applied to model transitions between Memory, Random Walk, and Mean-Reverting states, and their long-run destinations. Results show Bitcoin's full-period H (0.649) exceeded the S&P 500's (0.607) via our manual method, though Bitcoin's average H declined across most sliding-window eras (0.643 → 0.597 → 0.601 → 0.569), falling below the S&P 500's average (0.581) in the most recent recovery era. However, steady-state vectors reveal Bitcoin still spends 80.4% of its long-run time in the Memory state, an 18.4 percentage-point gap above the S&P 500's 62.0%. The evidence supports that Bitcoin's transition toward efficiency is significant but structurally incomplete, favoring the Fractal Market Hypothesis-that Bitcoin has transitioned toward, and temporarily beaten, real-world mature market efficiency-rather than full convergence under the Efficient Market Hypothesis.
The European Central Bank’s Appia initiative (March 2026) and its short-term precursor, Pontes, mark a structural reorganisation of European wholesale financial market infrastructure. This paper analyses Appia’s design and the distributed ledger technology (DLT) ecosystem emerging under the EU DLT Pilot Regime (Regulation 2022/858), and identifies two financial-stability vulnerabilities: the absence of a compliant automated secondary-market liquidity layer for tokenised instruments, and the dependency of European DLT infrastructure on non-European platforms, settlement stablecoins, and pricing oracles. Drawing on primary regulatory sources and the operational Spanish DLT ecosystem, and framed through network-externality and critical-infrastructure theory, it develops four analytical propositions: (i) infrastructure fragmentation may constitute a systemic risk structurally analogous to pre-TARGET2 settlement fragmentation; (ii) the absence of a MiFID II-compliant automated market-making (AMM) layer creates a structural liquidity trap; (iii) three critical components — settlement stablecoins, reference-rate oracles, and AMM governance — are currently controlled by US-domiciled entities, creating a ring-fenceable sovereignty gap; and (iv) Spain’s vertically integrated national ecosystem offers a replicable model for European financial stability. The paper proposes a seventh Appia building block — secondary-market liquidity infrastructure — and specifies the design of a compliant yield-space AMM, with implications for the Pontes pilot (Q3 2026) and the H2 2028 Appia blueprint. The propositions are theoretical and falsifiable but not yet empirically tested, pending operational Pontes data.
Huijun Dai, Sirui Wu, Min Han, Jing Li · 7 authors
Vision-Language Models (VLMs) have achieved remarkable progress in multimodal understanding tasks, but their high computational cost remains a significant barrier to practical deployment. Existing VLMs process all visual tokens equally regardless of input complexity, leading to substantial redundant computation. In this paper, we propose AdaToken-VLM, a novel adaptive token scheduling framework that dynamically allocates token budget based on input complexity. Our method consists of three key components: (1) a Task Complexity Estimator that analyzes question characteristics and visual attention patterns, (2) a Token Budget Predictor that maps complexity scores to appropriate token budgets with variance regularization, and (3) an Importance-Aware Token Selector that retains the most informative tokens without gradient computation. Extensive experiments on VQA v2 benchmark (214,354 validation samples) demonstrate that AdaToken-VLM achieves 50% token reduction with only 20.1% accuracy drop compared to the full LLaVA-1.5-7B model, while maintaining 86.3% accuracy at 75% token ratio. Our method reduces memory consumption by 25-50% and achieves up to 2.0× inference speedup, making VLMs more accessible for resource-constrained environments. Notably, our scheduler learns to differentiate complexity across question types (complexity std=0.1633), successfully adapting token allocation from 20 to 33 tokens based on input characteristics.
This study examines the influence of geopolitical risk (GPR) on cryptocurrency returns using daily global and monthly country-specific data from January 2019 to December 2025. Employing fixed effects panel regression, feasible generalised least squares, heteroskedasticity-based IV estimation, and panel quantile regression across eight major non-stablecoin cryptocurrencies, we find that GPR effects are asymmetric and regime-dependent GPR is associated with higher returns in bearish market conditions, remains insignificant under normal conditions, and dampens returns in bullish periods. GPR threats exert a consistently stronger negative effect than GPR acts across all market conditions, reflecting the forward-looking nature of investor behaviour in cryptocurrency markets. Country-specific analysis across 44 countries reveals a clear geographic divide: developed European economies including Sweden, Spain, and Finland show significant negative effects consistent with risk-off behaviour, while emerging and high crypto-adoption markets including South Korea, Saudi Arabia, and Vietnam show positive effects consistent with flight-to-crypto behaviour. Argentina shows a notable reversal between historical and recent GPR measures, reflecting its dramatic macroeconomic deterioration during the study period. Quantile-based Granger causality tests confirm that GPR has predictive power over future returns, particularly during bullish market conditions. These findings challenge the unconditional safe-haven characterisation of cryptocurrencies and carry direct implications for portfolio managers, regulators, and investors seeking safe-haven assets in an increasingly uncertain geopolitical environment.
We study the portfolios and characteristics of cryptocurrency owners, including their position in the wealth distribution. We first use Norwegian tax returns to investigate portfolios and find that 0.85 percent of the adult population declares owning cryptocurrency. We then use a battery of novel data sources and statistics to understand the extent of non-compliance, finding that the actual population of cryptocurrency owners was approximately 5 percent of the full population. We also find that cryptocurrency owners are notably younger than the typical owners of traditional types of assets, and that cryptocurrency has become a considerable wealth object among younger generations.
Bitcoin's base-layer throughput is bounded by its block interval and block-size limits, which constrains the rate at which individual transactions can be confirmed on-chain. This paper presents Uni-Speed Bridge, an off-chain transactionbatching framework that aggregates a set of pending transactions into a single, fixed-size cryptographic anchor using two complementary constructions: (i) a Merkle tree, which preserves per-transaction data availability and enables O(log n) inclusion proofs, and (ii) a modular polynomial evaluation over a large prime field, which serves as an auxiliary batch-level commitment. The system is implemented in Go and uses a bounded workerpool concurrency model to parallelize transaction hashing across available CPU cores, together with a write-ahead log for crash durability and a retrying, idempotent JSON-RPC client for interaction with a Bitcoin Core node. We describe the architecture, provide a complexity analysis of each stage, and are explicit about what the system does not provide: it does not modify Bitcoin consensus rules, does not itself validate transaction signatures, and has not undergone independent security audit or empirical benchmarking on production hardware. We position this work as an engineering case study in off-chain data-availability design rather than a validated scaling proof, and outline the concrete steps-signature validation, zero-knowledge succinctness proofs, and third-party audit-required before any production deployment.
Forced liquidations in cryptocurrency derivatives are almost always counted from exchange-published event feeds, in which one economic liquidation appears as several records. We quantify the resulting distortion on Hyperliquid, an on-chain perpetual-futures venue whose node-fills archive is complete and attributed to the liquidated account — a combination no centralised venue publishes. On a fixed-hour stratified sample of 351,540 liquidation episodes reconstructed from 2,010,042 fills over one year, counting fills rather than episodes inflates the event count by a factor of 5.72 (day-resampled 95% CI [5.47, 5.99]). The inflation is not a constant: it is size-dependent, rising from a median of 2 fills per episode below the median size to a median of 72 in the top percentile, so the bias is concentrated on precisely the events that liquidation studies are about. The top 1% of episodes generate 23.1% of all fills. We also characterise the episode-size distribution and report a negative result: the tail is unambiguously heavy — an exponential is rejected decisively — and it is not a power law, but it cannot be named beyond that. Lognormal and Weibull specifications both dominate a Pareto fit at every estimable threshold across nearly three orders of magnitude of cut-off, yet which of the two wins reverses with the threshold, so any named family would report the analyst's cut-off. Finally, the bias already documented in the literature runs the other way: rate-limited centralised-exchange feeds undercount, publishing per-second maxima rather than samples, while fill-level on-chain records overcount. Pooling the two sources combines a downward-biased count with an upward-biased one, and they do not cancel. All data, code and pre-registrations are public.
Nigeria’s land administration system embodies a structural paradox. It represents a robust legal framework that coexists with persistently unreliable administrative processes. Although land governance is anchored in the Land Use Act 1978 and supported by complementary state laws, the procedures for the formalisation and perfection of title remain slow, opaque, and susceptible to manipulation. These inefficiencies impose significant economic costs and continue to constrain the productive potential of a rapidly expanding real estate sector. Crucially, these challenges arise not from deficiencies in the law itself, but from the operational weaknesses of the institutional framework responsible for its implementation. Over the past two decades, reform efforts, most notably, the Abuja Geographic Information System (AGIS) and Lagos Land Information Management System (LIMS) (later named the “Lagos e-GIS” in 2024), have introduced digitisation into land administration processes. While these initiatives have improved record management and reduced certain procedural delays, they have not addressed the deeper structural problem, that is, the absence of a reliable trust architecture capable of ensuring the integrity, transparency, and accountability of administrative actions. This paper argues that Nigeria’s land registry failures are fundamentally institutional rather than technological. It proposes a hybrid permissioned blockchain framework as an additional administrative layer within the existing legal system. Unlike conventional digitisation, which alters the form of record-keeping without addressing control and verification, the proposed framework introduces cryptographic accountability into the execution of administrative functions. Through the use of sequential, multi-signature smart contracts and a verifiable distributed ledger, each stage of the title perfection process becomes transparent, auditable, and resistant to unauthorised alteration. Drawing on comparative insights from jurisdictions including Georgia, Sweden, Estonia, Ghana, Kenya, and Namibia, the paper develops a feature-to-failure mapping that links blockchain’s operational capabilities to specific deficiencies within Nigeria’s land administration system. It concludes that blockchain should not replace existing institutions but should instead reinforce them by embedding verifiability, accountability, and procedural integrity into the exercise of administrative authority.
The paper analyzes the potential of blockchain technology for managing and protecting aircraft maintenance records in maintenance, repair and overhaul (MRO) and continuing airworthiness management (CAMO) organizations under the European Union Aviation Safety Agency (EASA) regulatory framework. Based on a review of scientific literature, EASA regulations and guidelines on electronic records and signatures, and a case study of the combined airworthiness organization SK.CAO.003, the paper derives functional and security requirements for an information system handling maintenance data. These requirements, which concern in particular data integrity, traceability, redundancy, availability, authorization and auditability, are compared with the properties of public, private and consortium blockchain networks. The paper proposes a concept of a permissioned consortium blockchain architecture based on Hyperledger Fabric, using distributed ledger technology for storing immutable metadata and cryptographic hashes of maintenance records while keeping detailed documents in existing off-chain MRO systems. The concept is illustrated by two web-based proof-of-concept applications: the first records aircraft-level maintenance events and visualizes chain immutability, the second tracks the life cycle of individual components from registration to retirement. The results indicate that blockchain can meet key regulatory requirements for maintenance records management, while identified limitations relate mainly to integration with legacy systems, identity and access management, and operating costs of the network.
Md Al Amin, Indrajit Ray, Indrakshi Ray, Yashwant K. Malaiya · 5 authors
Access to electronic health records (EHRs) is heavily regulated by various policies, including federal-level policies, state-level statutes, international data protection laws, and local and organizational-level policies. These policies may include procedures to ensure compliance with other organizational-level regulations. In addition, individual patients can establish agreements, formally known as patient-provider agreements (PPA), with their healthcare providers to express their consent to access or share their protected health information (PHI). When such policies are adequately specified and implemented, they go a long way toward protecting EHR data. However, research has shown that significant policy compliance problems or gaps often go undetected until after a breach or security incident. Further, a recent study shows that subcultures within a healthcare organization influence whether employees violate policies, perhaps unintentionally. These observations motivate us to revisit the compliance and provenance aspects of policies. This dissertation proposes a blockchain-powered, smart contract-based policy-compliance assurance framework to enforce patient-provider agreements and other applicable policies and attributes, ensuring policy compliance and provenance in the healthcare sector. This work proposes a novel compliance review mechanism, Proof of Compliance (PoC), that conducts reviews through a set of independent, distributed, decentralized auditor nodes from various stakeholders, such as healthcare organizations, insurance companies, federal and other government agencies, regulatory agencies, and others mandated by the business requirements. Blockchain smart contracts appear to be a promising new technology for enforcing policies. In addition, blockchains' immutable storage properties and strong integrity guarantees provide hope that an adequate trail of policy compliance (or non-compliance) can be maintained, thereby facilitating provenance.
We prove that first homology of the control flow graph provides a complete characterization of reentrancy vulnerability in smart contracts. Specifically, we establish the Homological Reentrancy Theorem: a contract admits a reentrant execution path if and only if H₁(G) ≠ 0, where G is the extended control flow graph incorporating external call returns. We prove soundness (no false negatives) and completeness (no false positives) for contracts satisfying a non-degeneracy condition. For multi-contract systems, we apply the Mayer-Vietoris exact sequence to compute H₁ of the composed system from individual components, enabling detection of cross-contract reentrancy. We validate empirically against 17 known exploits including The DAO (2016), Parity Wallet (2017), and Cream Finance (2021), achieving 100% detection with zero false positives.