arXiv subjects

Haoqian Zhang

Publications and source records attributed to Haoqian Zhang.

10 recordsLinked to original sources

arXiv2026

AISPA: User-Centric System Prompt Auditing for Large Language Model Applications

System prompts are instructions configured by developers to govern the behaviors of foundation models in AI applications. They are used throughout commercial AI products, but are rarely disclosed to the public or regulators, creating a serious trust and accountability gap in the wide deployment of AI systems. In this paper, we introduce Artificial Intelligence System Prompt Assurance (AISPA), a user-centric framework for systematically auditing system prompts in AI systems. AISPA examines specific parts of a system prompt and evaluates them along eight dimensions that matter to users. We then use this framework to review 3,249 instructions from system prompts in 88 commercial AI products, classifying each instruction as either protective (of users) or problematic. Our audit surfaces four core findings. First, system prompt design varies substantially across products and developers, with some organizations averaging over 60 protective instructions per product while others average fewer than 5. Second, protective instructions are widely adopted but shallow in scope: 98.9% of products contain at least one, yet only 24% cover all eight dimensions of the AISPA taxonomy. Third, system prompts have grown steadily longer and more protective of users, suggesting that user protection is becoming a more visible concern in commercial prompt design. Fourth, despite this progress, problematic instructions remain pervasive: roughly 40% of products contain at least one instruction that works against user interests, and protective and problematic instructions frequently coexist within the same prompt. Our findings highlight the need for greater transparency, standardization, and independent oversight for system prompts in commercial AI products.

cs.AI↗

arXiv2026

Humanly: A Configurable and Traceable Environment for Human-AI Collaborative Writing

Teachers, conference chairs, and public readers all judge writing from limited evidence, seeing only a finished document and not the process that produced it. Final text alone cannot reveal whether a document was produced through human typing, AI generation, or mixed human-AI collaboration. Existing process-tracking tools help, but many are tied to host-document histories, provide coarse activity records, and offer limited control over the writing environment. Humanly is a writing platform that makes the writing process itself the evidence. Users configure writing environments for personal documents or assigned tasks and draft in a workspace that records writing activity and in-platform AI assistance. Humanly can package a completed session into a sealed writing certificate with configuration-aware anomaly behavior review. It can support writing scenarios such as course assignments, peer review, and personal certification. Our user study shows that Humanly is helpful across roles, and a red-teaming study shows that the Humanly Typing Detector distinguishes human hand typing from automated typing.

cs.CL↗

arXiv2026

PRISM: Refracting the Entangled User Behavior Space for E-Commerce Search

E-commerce search systems rely on modeling user behavior to estimate item relevance and user preference, which are typically assumed to be stable and independently learnable signals. However, in practice, user interactions are jointly shaped by exposure mechanisms, feedback loops, and semantic matching, leading to entangled and dynamically drifting behavioral signals. As a result, both preference estimation and relevance modeling suffer from confounding effects and semantic misalignment, which limits the robustness of downstream ranking models. To address this issue, we propose PRISM, a Preference-Relevance Interaction Semantic Modeling framework for e-commerce search behavior prediction. PRISM explicitly models the interaction between user preference and item relevance rather than treating them as independent components. Specifically, it introduces a preference rectification module to iteratively refine user preference under relevance-aware constraints, improving robustness against behavioral confounding. To ensure semantic consistency, we further incorporate a large language model (LLM)-driven semantic anchoring mechanism that leverages positive and negative prototypes to calibrate relevance representations. Finally, a preference-conditioned evidence routing module adaptively aggregates multi-source behavioral signals, enabling context-aware and preference-aligned relevance estimation. Extensive experiments on two public e-commerce benchmarks demonstrate that PRISM consistently outperforms strong baselines, validating the effectiveness of explicitly modeling preference-relevance interaction for robust and semantically grounded search behavior modeling.

cs.IR↗

arXiv2025

TRIP: Coercion-resistant Registration for E-Voting with Verifiability and Usability in Votegral

Online voting is convenient and flexible, but amplifies the risks of voter coercion and vote buying. One promising mitigation strategy enables voters to give a coercer fake voting credentials, which silently cast votes that do not count. Current systems along these lines make problematic assumptions about credential issuance, however, such as strong trust in a registrar and/or in voter-controlled hardware, or expecting voters to interact with multiple registrars. Votegral is the first coercion-resistant voting architecture that leverages the physical security of in-person registration to address these credential-issuance challenges, amortizing the convenience costs of in-person registration by reusing credentials across successive elections. Votegral's registration component, TRIP, gives voters a kiosk in a privacy booth with which to print real and fake credentials on paper, eliminating dependence on trusted hardware in credential issuance. The voter learns and can verify in the privacy booth which credential is real, but real and fake credentials thereafter appear indistinguishable to others. Only voters actually under coercion, a hopefully-rare case, need to trust the kiosk. To achieve verifiability, each paper credential encodes an interactive zero-knowledge proof, which is sound in real credentials but unsound in fake credentials. Voters observe the difference in the order of printing steps, but need not understand the technical details. Experimental results with our prototype suggest that Votegral is practical and sufficiently scalable for real-world elections. User-visible latency of credential issuance in TRIP is at most 19.7 seconds even on resource-constrained kiosk hardware. A companion usability study indicates that TRIP's usability is competitive with other e-voting systems, and formal proofs support TRIP's combination of coercion-resistance and verifiability.

cs.CR↗

arXiv2025

Rational Censorship Attack: Breaking Blockchain with a Blackboard

Censorship resilience is a fundamental assumption underlying the security of blockchain protocols. Additionally, the analysis of blockchain security from an economic and game theoretic perspective has been growing in popularity in recent years. In this work, we present a surprising rational censorship attack on blockchain censorship resilience when we adopt the analysis of blockchain security from a game theoretic lens and assume all users are rational. In our attack, a colluding group with sufficient voting power censors the remainder nodes such that the group alone can gain all the rewards from maintaining the blockchain. We show that if nodes are rational, coordinating this attack just requires a public read and write blackboard and we formally model the attack using a game theoretic framework. Furthermore, we note that to ensure the success of the attack, nodes need to know the total true voting power held by the colluding group. We prove that the strategy to join the rational censorship attack and also for nodes to honestly declare their power is a subgame perfect equilibrium in the corresponding extensive form game induced by our attack. Finally, we discuss the implications of the attack on blockchain users and protocol designers as well as some potential countermeasures.

cs.GT↗

arXiv2024

E-Vote Your Conscience: Perceptions of Coercion and Vote Buying, and the Usability of Fake Credentials in Online Voting

Online voting is attractive for convenience and accessibility, but is more susceptible to voter coercion and vote buying than in-person voting. One mitigation is to give voters fake voting credentials that they can yield to a coercer. Fake credentials appear identical to real ones, but cast votes that are silently omitted from the final tally. An important unanswered question is how ordinary voters perceive such a mitigation: whether they could understand and use fake credentials, and whether the coercion risks justify the costs of mitigation. We present the first systematic study of these questions, involving 150 diverse individuals in Boston, Massachusetts. All participants "registered" and "voted" in a mock election: 120 were exposed to coercion resistance via fake credentials, the rest forming a control group. Of the 120 participants exposed to fake credentials, 96% understood their use. 53% reported that they would create fake credentials in a real-world voting scenario, given the opportunity. 10% mistakenly voted with a fake credential, however. 22% reported either personal experience with or direct knowledge of coercion or vote-buying incidents. These latter participants rated the coercion-resistant system essentially as trustworthy as in-person voting via hand-marked paper ballots. Of the 150 total participants to use the system, 87% successfully created their credentials without assistance; 83% both successfully created and properly used their credentials. Participants give a System Usability Scale score of 70.4, which is slightly above the industry's average score of 68. Our findings appear to support the importance of the coercion problem in general, and the promise of fake credentials as a possible mitigation, but user error rates remain an important usability challenge for future work.

cs.HC↗

arXiv2023

F3B: A Low-Overhead Blockchain Architecture with Per-Transaction Front-Running Protection

Front-running attacks, which benefit from advanced knowledge of pending transactions, have proliferated in the blockchain space since the emergence of decentralized finance. Front-running causes devastating losses to honest participants and continues to endanger the fairness of the ecosystem. We present Flash Freezing Flash Boys (F3B), a blockchain architecture that addresses front-running attacks by using threshold cryptography. In F3B, a user generates a symmetric key to encrypt their transaction, and once the underlying consensus layer has finalized the transaction, a decentralized secret-management committee reveals this key. F3B mitigates front-running attacks because, before the consensus group finalizes it, an adversary can no longer read the content of a transaction, thus preventing the adversary from benefiting from advanced knowledge of pending transactions. Unlike other mitigation systems, F3B properly ensures that all unfinalized transactions, even with significant delays, remain private by adopting per-transaction protection. Furthermore, F3B addresses front-running at the execution layer; thus, our solution is agnostic to the underlying consensus algorithm and compatible with existing smart contracts. We evaluated F3B on Ethereum with a modified execution layer and found only a negligible (0.026%) increase in transaction latency, specifically due to running threshold decryption with a 128-member secret-management committee after a transaction is finalized; this indicates that F3B is both practical and low-cost.

cs.CR↗

arXiv2023

Breaking Blockchain Rationality with Out-of-Band Collusion

Blockchain systems often rely on rationality assumptions for their security, expecting that nodes are motivated to maximize their profits. These systems thus design their protocols to incentivize nodes to execute the honest protocol but fail to consider out-of-band collusion. Existing works analyzing rationality assumptions are limited in their scope, either by focusing on a specific protocol or relying on non-existing financial instruments. We propose a general rational attack on rationality by leveraging an external channel that incentivizes nodes to collude against the honest protocol. Our approach involves an attacker creating an out-of-band bribery smart contract to motivate nodes to double-spend their transactions in exchange for shares in the attacker's profits. We provide a game theory model to prove that any rational node is incentivized to follow the malicious protocol. We discuss our approach to attacking the Bitcoin and Ethereum blockchains, demonstrating that irrational behavior can be rational in real-world blockchain systems when analyzing rationality in a larger ecosystem. We conclude that rational assumptions only appear to make the system more secure and offer a false sense of security under the flawed analysis.

cs.GT↗

arXiv2020

Economic Principles of PoPCoin, a Democratic Time-based Cryptocurrency

While democracy is founded on the principle of equal opportunity to manage our lives and pursue our fortunes, the forms of money we have inherited from millenia of evolution has brought us to an unsustainable dead-end of exploding inequality. PoPCoin proposes to leverage the unique historical opportunities that digital cryptocurrencies present for a "clean-slate" redesign of money, in particular around long-term equitability and sustainability, rather than solely stability, as our primary goals. We develop and analyze a monetary policy for PoPCoin that embodies these equitability goals in two basic rules that maybe summarized as supporting equal opportunity in "space" and "time": the first by regularly distributing new money equally to all participants much like a basic income, the second by holding the aggregate value of these distributions to a constant and non-diminishing portion of total money supply through demurrage. Through preliminary economic analysis, we find that these rules in combination yield a unique form of money with numerous intriguing and promising properties, such as a quantifiable and provable upper bound on monetary inequality, a natural "early adopter's reward" that could incentivize rapid growth while tapering off as participation saturates, resistance to the risk of deflationary spirals, and migration incentives opposite those created by conventional basic incomes.

cs.CY↗

arXiv2020

Infnote: A Decentralized Information Sharing Platform Based on Blockchain

Internet censorship has been implemented in several countries to prevent citizens from accessing information and to suppress discussion of specific topics. This paper presents Infnote, a platform that helps eliminate the problem of sharing content in these censorship regimes. Infnote is a decentralized information sharing system based on blockchain and peer-to-peer network, aiming to provide an easy-to-use medium for users to share their thoughts, insights and views freely without worrying about data tampering and data loss. Infnote provides a solution that is able to work on any level of Internet censorship. Infnote uses multi-chains architecture to support various independent applications or different functions in an application.

cs.CR↗