The transition to heavy-duty battery electric vehicles requires an efficient and cost-effective deployment of the charging infrastructure, particularly when multiple operators share resources. This paper presents a multi-phase optimization framework for the joint planning of charging stations in a shared network, using high-resolution empirical truck trajectory data from two freight companies with distinct operational characteristics. The model is formulated to minimize the total number of charging stations while ensuring that the predefined electrification targets are met over successive expansion stages. The analysis captures heterogeneity in fleet usage, with one company operating a spatially concentrated network with shorter and more consistent routes, and the other exhibiting more dispersed operations with longer and more variable driving patterns. The results show that early-stage infrastructure deployment primarily supports fleets with concentrated operations, while later expansion phases are essential to accommodate long-haul and geographically dispersed transport demand. Furthermore, shared infrastructure not only enables reductions in redundant investments, but also introduces dependencies where certain fleets rely heavily on the full network to sustain electrified operations. In general, the findings highlight the importance of coordinated and data-driven infrastructure planning, and demonstrate that fleet-specific characteristics strongly influence both infrastructure requirements and electrification outcomes. The proposed framework provides practical insights on how collaborative and phased deployment strategies can enhance the scalability and efficiency of freight transport electrification.

As energy prices surge for the second time in recent years driven by the ongoing crisis in the Middle East, the European Union's continuing reliance on fossil energy imports is becoming increasingly apparent. However, despite offering an intriguing prospect of improved energy resilience, the ramp-up of local green hydrogen production lags far behind the officially stated ambitions set after the 2022 energy crisis. A prominent reason for the widening implementation gap between announced and realised production projects is overly strict rules on renewable power sourcing, prompting Member states' ministries and the European Commission to propose advancing a planned rules review from 2028 to 2026. To contribute to a successful review and rule adjustments, we address an important gap in understanding the effects of power purchase rules on green hydrogen production. By taking the perspective of European electrolyser operators, we show how the criterion of additionality and its interaction with required temporal correlation can jeopardise the fulfilment of green hydrogen offtake agreements and affect green hydrogen production costs across different European bidding zones. Applying different design paradigms to a green hydrogen production system reveals that electrolyser operator measures, such as PPA and storage upsizing, can help to mitigate the business risks posed by the additionality criterion but come with increased costs. Alternatively, relaxed temporal correlation and increased offtake flexibility both increase production system robustness and reduce production costs simultaneously. Whereby relaxing temporal correlation rules does not result in exceeded emission intensity thresholds, underlining the potential of extended transitional rules to support the ramp-up of European green hydrogen production.

Unmanned aerial vehicle (UAV) relays deliver flexible, on-demand wireless coverage, but jointly tuning the position, altitude, transmit power and bandwidth of the relay is a non-convex, heavily constrained optimization task that easily traps swarm-based optimizers in poor local optima. We propose PWOA, a Polynomial-decay and Pinhole-imaging Whale Optimization Algorithm with three complementary improvements: (i) a Good Nodes Set (GNS) initialization that spreads the initial population uniformly across the search space; (ii) a polynomial nonlinear schedule for the convergence factor that prolongs early exploration and sharpens late exploitation; and (iii) a stagnation-triggered pinhole-imaging opposition-based learning (POBL) operator paired with an elite Gaussian local search, which together escape local optima while refining the leader. On a five-dimensional UAV relay deployment problem with five inequality constraints ($N{=}30$, $T{=}500$, 30 independent runs), PWOA simultaneously attains the lowest Best, Worst, Mean and standard deviation among PWOA, WOA, SCA and IPSO, cutting the mean by $1.4$--$18.5\%$ and the standard deviation by $15$--$87\%$ over the three baselines, and exhibits the fastest average convergence.

Information retrieval systems have long treated semantic similarity as a proxy for relevance. For constraint-sensitive queries, this proxy can fail when a document is topically close to the query but supports the opposite constraint direction, such as satisfying an attribute that should be excluded or affirming a relation that should be negated. We study this failure as constraint-violating evidence exposure and propose CoDeR, a local constraint-compatible dense retrieval method that separates topical relevance from constraint compatibility. CoDeR keeps a standard topical encoder for candidate coverage and adds a compatibility scorer, implemented as a bi-encoder, trained with lexical-polarity supervision over contrastive satisfying and violating evidences. The compatibility signal can be used to rescore topical candidates or to retrieve an auxiliary compatibility-oriented candidate set, producing a ranked document list without external Large Language Model~(LLM) calls at inference time. We evaluate CoDeR on controlled diagnostics and public negative-constraint retrieval benchmarks. Across three controlled diagnostic sets targeting antonymy, negation, and exclusion, CoDeR reduces V@2 by 20.59, 23.53, and 5.77 points relative to the strongest non-CoDeR baselines, and improves FVR by pushing the first violating document deeper in the ranking.

In this work we settle the complexity of three sketching problems. (i) We show that sketching vertex neighborhood sizes in graphs requires $Ω(n^2)$ bits, standing in sharp contrast to the $\tilde{O}(n)$ complexity of sketching edge cuts. (ii) We obtain tight lower and upper bounds of $\tildeΘ(n^2)$ for sketching coverage functions with additive and multiplicative errors. (iii) We prove an $Ω(n^2)$ lower bound for sketching Random Utility Models under the $\ell_\infty$-norm, improving upon the previous $Ω(n \log n)$ bound and matching a known upper bound to within logarithmic factors. These bounds are obtained through a connection with the problem of sketching the intersection profile of a distribution $D$ on $2^{[n]}$. Specifically, we seek a succinct data structure that, for any query set $S \subseteq [n]$, approximates the quantity $\Pr_{T \sim D}[T \cap S \neq \varnothing]$ to within a small constant additive error. One can obtain lower bounds for this latter problem directly from known results about the itemset frequency estimation problem in databases for which tight bounds are known. As an additional contribution, we also provide an alternative proof for the intersection profile sketching lower bound, in the setting in which the accuracy parameter is constant. This proof relies solely on elementary probability avoiding the heavier machinery used in previous proofs.

Federated Learning(FL) is predominantly deployed in enterprise environments, where limited transparency and restricted auditability hinder broader adoption. Existing FL systems often suffer from opaque aggregation processes, making it unclear which model updates are accepted or discarded. Current mitigation strategies typically rely on external validators introducing additional computational and communication overhead. In this paper, we propose a novel FL framework that leverages existing Web3 technologies to enhance transparency, trust and auditability throughout the training process. The framework adopts a hierarchical architecture in which delegated managers orchestrate the FL training process within their respective federations. To mitigate adversarial and poisoning attacks, a combination of novelty detection and consensus mechanisms were employed. Model updates are encoded and broad casted to all managers, who independently evaluate their validity and those model updates that are approved by the consensus are incorporated into the global model. Additionally, a reputation score based backup mechanism is employed to ensure model generation. Extensive experiments conducted under real world scenarios demonstrate the effectiveness, resilience of the proposed framework, highlighting its potential to enable transparent FL beyond traditional enterprise setting.

Code review has been the primary quality gate in software development since Fagan formalised code inspection in 1976. For five decades, having a human examine and comment on a colleague's changes before merge has been a cornerstone practice at organisations of every size. Coding agents are large language model (LLM)-based autonomous systems capable of reading, writing, testing, and repairing software. We argue that coding agents have crossed a threshold of capability at which traditional human code review is no longer a necessary component of a software quality pipeline. Our argument rests on two claims: every stated goal of code review can be served by agents at lower cost and higher throughput; the naive integration in which agents write code and humans remain the mandatory reviewers is a dead end because it neither provides meaningful assurance nor scales with AI-assisted throughput.

In this work, we provide a probabilistic derivation of a class of multiparameter signless Stirling numbers of the first kind and their q-analogues, and study the associated multivariate discrete distributions.

We present an algorithmic framework for the exhaustive generation and tabulation of knot and link diagrams on the thickened torus T^2 x I, based on the theory of maps on surfaces. Cellular 4-regular torus projections are encoded by permutation pairs (alpha, sigma), and unsensed equivalence classes are enumerated completely and without duplication via canonical representatives. Crossing assignments, local diagram-level reductions, and the generalized Kauffman-type bracket are formulated entirely within the same permutation model. The pipeline is validated against published genus-one classifications for crossing numbers N <= 5 and then extended to N = 6, 7, 8, producing, to our knowledge, the first complete genus-one tabulation at these crossing numbers under the stated comparison conventions. The resulting dataset contains more than 33,000 knot and link types. Besides the tables, the computation yields proved structural facts, including a parity statement for the a-span of the bracket and a sharp upper bound N-1 for the number of bigon faces in a 4-regular torus map. It also suggests several conjectures, among them a formula for the maximum number of straight-ahead components, the absence of equi-quadrilateral knot projections, and a 4N upper bound for the genus-one bracket span.

Permutations avoiding a pattern of length three are enumerated by the Catalan numbers. In this work, we present methods for ranking and unranking such permutations in lexicographic or colexicographic order.

The degree of convexity of a convex polyomino P is the smallest integer k such that any two cells of P can be joined by a monotone path inside P with at most k changes of direction. In this paper, we present a set of formulas and equations that are the basis of a C++ program that allows you to compute the longest counting sequence known to date (with respect to the area) of convex polyominoes of degree of convexity at most 2

Given a discrete rectangle R of dimensions h x w, let W be the set of snake-like polyominoes contained in R represented as binary matrices, i.e. polyominoes whose underlying simple graph is a chain with respect to the 4-adjacency relation. We present an algorithm that generates W for any h and w. Also, let a be the maximal area that can be realized by an element of W. We provide exact formulas of a for h <= 5 and any w.

We study Fibonacci (staircase) polyominoes, a class of column-convex polyominoes whose lower boundary is a staircase with unit vertical steps. We derive multivariate generating functions that refine Turban's Fibonacci-number enumeration by tracking additional perimeter and area parameters. The proofs use a catalytic functional equation and, in a perimeter specialization, the kernel method, leading to explicit closed forms and Catalan-number coefficient formulas.

We study k-coloured Motzkin paths, namely Motzkin paths in which horizontal steps can be coloured in k different ways, and investigate their connection with the number of prefixes ending at odd height from both an analytical and a combinatorial point of view. Moreover, the combinatorial approach provides a random generation algorithm for k-coloured Motzkin paths in linear-time.

RedNote (a.k.a., Xiaohongshu, a global-scale social network platform) widely adopts approximate nearest neighbor search (ANNS) to power its search, recommendation, and advertising services. Due to the demanding Service Level Agreements (SLAs), we have to rely on in-memory graph-based ANNS (i.e., HNSW) to provide high throughput and low latency. However, the ever-growing user base and content volume have led to an explosive increase in memory footprint and consequently huge CapEx and OpEx. After exploring various alternatives, we find that building a clustering-based ANNS on top of all-flash servers can be promising. Yet, we still experience severe overheads from the kernel I/O stack, a fixed pruning strategy, and slow index construction. We present HELMSMAN, a high-performance and cost-effective clustering-based ANNS system, which combines an ANNS-oriented userspace storage stack, a leveling-learned pruning module, and GPU-accelerated pipelines of construction. HELMSMAN saves over 90% of hardware costs and enables billion-scale index (re)builds within hours. In the current production deployment, operating stably for several months, 40 machines now host ANNS workloads that previously required about 35,000 cores and 0.35 PB DRAM.

The LIMITS community was founded to foster conversations that move away from growth-oriented visions and values in computing toward a focus on long-term well-being. This orientation, we argue, inherently engages questions of time and temporality. Prior work has shown that temporal frameworks shape how futures are imagined, which problems are understood to be worth attending to, and which solutions or alternatives are pursued. We begin this paper with author observations of time in their lived experience, and then extend these observations to the LIMITS community. Through a systematic literature review of the last decade of LIMITS scholarship, we identify ways that explicit attention to how concepts of time and temporality are understood would enrich Limits scholarship. Within the LIMITS scholarship that does engage with time, we identify five recurring types of temporal engagement: computing time, methodological and design time, politics and ethics of time, biological and ecological time, and afterlife and waste time. Together, these engagement types highlight how implicit assumptions about time are embedded across research practices, design approaches, and accounts of technological impact within LIMITS work. We discuss these findings in relation to cross-disciplinary scholarship that takes time as an analytic concern and consider how these patterns point to a broader need for more explicit, plural, and situated engagements with time in the LIMITS community, and why this matters for the community's commitments.

Proxies, VPNs and Tor have long helped the privacy community and users in censored regions to fight censorship. However, the same tools can be maliciously exploited by malware and botnets to conceal their communication to external command and control servers. Despite being a critical concern fueled by the proliferation of malware based attacks, no longitudinal studies have analyzed how malware applications use covert channels (CC) to evade detection. We fill this gap by performing the first study of the usage of covert channels in the Android malware ecosystem. To that end, we develop a multistage pipeline that combines static and dynamic analysis to investigate both system and network-level features. We applied this pipeline on a corpus of 3.5M Android malware spanning 2009 to July 2025. Our carefully crafted static validation rules uncovered 288K APKs that used CCs spanning 511 malware families and CC usage growing exponentially from 0.30\% (2012) to 50\% (2025). Overall, in dynamic analysis, we identified 19,308 unique IP addresses being contacted in 85 countries, out of which we were able to explicitly validate the presence of CCs for 59 IP addresses across 17 countries. Further, we performed a longitudinal dataset study spanning over 16 years for CC based malware and found that CC usage has evolved, \textit{e.g.,} some malware adopted by using more than one CCs; others switched between them periodically (one family switched CC usage 40 times from 2019 to 2025).

Extensive-form games (EFGs) are a standard model for sequential decision-making in games. A fundamental and typically implicit assumption in EFGs is that players always have access to all of their actions at every decision point. However, in many realistic settings, certain actions might be unavailable during game-play due to exogenous stochasticity, hindering the expressivity of the standard EFG model. Given a `base' EFG, we formalize a model that allows for actions to be stochastically restricted, leading to a corresponding Extensive-Form Games with Stochastic Action Sets (EFGSAS). In EFGSAS, we derive an expansion procedure that results in an equivalent EFG, thus showing that standard strategy formalisms could require exponentially-large representations. However, under an appropriate independence assumption, we show that compact strategy representations polynomial in the size of the base EFG exist. Computationally, we introduce an algorithm called SI-CFR that minimizes sleeping internal regret, converging to Nash equilibria with high probability in two-player zero-sum EFGSAS. Finally, we utilize a stochastic approximation procedure to recover compact representations of Nash equilibria, utilizing only the iterates of SI-CFR.

Approximate multipliers deliberately relax computational accuracy to achieve gains in power efficiency, latency, and silicon area, which makes them well-suited for error-resilient applications such as neural networks. In this work, we introduce a co-evolutionary algorithm that leverages an off-the-shelf large language model (LLM) without requiring domain-specific training to automate the design of optimized 8-bit approximate multipliers. The approach simultaneously evolves a population of candidate circuits and a population of prompt templates that steer LLM-driven modifications. Experimental results for several target design objectives demonstrate that the proposed method discovers approximate multipliers with improved error-area trade-offs compared to highly optimized circuits from the EvoApproxLib library.

Space data centers (SDCs) are emerging as a promising orbital computing infrastructure for the future AI industry. Unlike conventional satellites that mainly serve as relay nodes or lightweight onboard processors, SDCs integrate communication, computing, storage, and control capabilities in orbit, enabling persistent service support for data-intensive and intelligence-driven space applications. In this article, we investigate how SDCs may transform space communication paradigms from connectivity-oriented data transmission toward task-oriented and service-centric information exchange. We first present a hierarchical SDC network architecture consisting of access, relay, computing, and control layers, and outline possible deployment strategies. We then explore representative future application scenarios enabled by SDCs, highlighting their communication characteristics and associated research challenges. Simulation results further demonstrate the effectiveness of SDCs in reducing control-layer latency in hierarchical space networks. Finally, we identify key research directions toward the practical deployment of SDCs.

We study the problem of allocating indivisible goods under constraints, expressed via a conflict graph $G$. In such an instance, the $m$ items are the vertices of $G$ and connected items cannot be allocated in the same bundle. Under this model, it is already known that EF1 allocations may not exist. Our main contribution is an analysis parametrized by the maximum degree $Δ(G)=Δ$ on the existence and computation of complete EF1 allocations. We address this question in various cases by leveraging results from matching theory. First, we provide a tight existence result for agents with ordered valuations and for the broader class of tiered valuations. We present an algorithm that returns an EF1 allocation when then number of items does not exceed a specific bound. This bound is determined by $n$ and $Δ$, and it is tight when $Δ$ is greater than $2n/3$. We also construct an approximation algorithm when $m$ exceeds this bound. For general additive valuations the problem becomes more challenging. Given the current impossibility results, we focus on the case where the number of items is at most $2n$. For this case, we provide an almost complete picture for the instances that admit EF1 allocations, by combining Round Robin with matchings.

Future 6G networks will operate across distributed and heterogeneous domain infrastructures, making conventional single-domain management insufficient for proactive, trustworthy automation. Network Digital Twins (NDTs) enable what-if analysis, AI-assisted optimization, and risk-free validation of control actions before deployment, yet monolithic end-to-end twins remain impractical due to scalability, fidelity, and cross-domain coordination challenges. Accordingly, this paper proposes a Digital Twin-enabled 6G architecture that exposes NDT capabilities as a specialized service domain within a multi-domain orchestration framework built on a state-of-the-art service-based 6G architecture. A DT Orchestrator interprets \textit{predictive} and \textit{prescriptive} what-if queries and composes domain-specific DT modules and simulators on demand, while decision authority remains with the requesting entity. Furthermore, a generalized workflow covers telemetry synchronization, simulation-based decision support, and closed-loop execution. The framework is demonstrated through a green-networking use case that couples a system-level O-RAN cellular digital twin component with a two-stage solar-allocation simulator, evaluated over a 105-base-station deployment in Poznan using simulative datasets. Joint coverage and renewable optimization reduces daily grid consumption by 28.5\% with 32 solar panels at the diminishing-returns threshold, with 17 base stations identified as both coverage-active and high-priority solar candidates as evidence that cross-domain NDT coordination enables sustainable, intent-driven 6G network management.

Maximin share (MMS) stands out as a central notion in fair resource allocation. It is known that exact MMS fairness is not always attainable, especially when agents differ along two dimensions: their valuations and their perceptions of the divisibility of resources. The former case with heterogeneous valuations has been widely studied in the literature. The latter, referred to as subjective divisibility by Bei et al., [Games Econ. Behav. 2025], remains much less explored. We study MMS approximation under subjective divisibility. First, we prove that even in the unary valuation setting, where all items have equal value, the optimal approximation ratio is 2/3. This result is somewhat surprising since in the objective setting, even when agents have heterogeneous valuations, the best possible approximation ratio is at least 7/9 [Huang and Zhou, 2025]. We then address the general case with both valuation heterogeneity and subjective divisibility. Previous work shows the existence of a 1/2-approximate MMS allocation. In this paper, we develop new algorithmic techniques that overcome the difficulties posed by subjective divisibility, and improve the approximation guarantee to 5/9. Finally, we complement this result with small-agent cases. For up to four agents, we give polynomial-time algorithms that compute 2/3-approximate MMS fair allocations. These bounds are tight. Our results deepen the understanding of MMS fairness under heterogeneous valuations and subjective divisibility, and provide a new perspective for this emerging model.

One-day vulnerabilities pose significant risks due to delayed or incomplete patch adoption. Generating proof-of-concept (PoC) inputs is therefore essential for assessing real-world impact. The key challenge is identifying necessary constraints for triggering the vulnerability and solving them effectively. Existing directed fuzzing approaches prioritize inputs toward target locations, but neither explicitly identify necessary constraints nor solve them effectively, relying instead on target-distance feedback and random mutation. Agentic approaches show strong potential through code reasoning and structured input generation, but goal drift in long-horizon reasoning limits their effectiveness. DIG addresses this challenge by exploiting a key property of one-day vulnerabilities: patches often reveal necessary preconditions for triggering. DIG uses an LLM to analyze the patch and synthesize an oracle making these conditions explicit. The oracle supports effective PoC generation at two levels. At the high level, DIG performs oracle-guided generator evolution, where an agent infers and solves constraints to satisfy the oracle. At the low level, DIG instruments the oracle into the target program and uses branch-distance feedback to guide random mutation in directed fuzzing. Evaluation shows DIG outperforms 2 state-of-the-art agents and 10 fuzzers across 138 real-world CVEs. DIG triggers 80 vulnerabilities, surpassing prior results and outperforming the best baseline by 40% (57 vs. 80 CVEs). Notably, DIG exclusively triggers 9 vulnerabilities no existing technique can trigger. Compared to the average of other tools, DIG triggers vulnerabilities faster in 92.9% of cases, achieving over 100x speedup in 48.8% of cases, with a maximum speedup of 3,664x. Beyond one-day PoC generation, DIG uncovers 6 previously unknown vulnerabilities in widely deployed libraries, enabling zero-day discovery.

Personalized outfit recommendation poses a significant challenge in e-commerce and social media platforms, requiring systems that balance user preferences with aesthetic compatibility. Collaborative filtering (CF) provides a traditional solution for this, but it struggles with data-sparse scenarios and complex user-item-outfit relationships. Meanwhile, existing template-based approaches are constrained by rigid pre-designed structures. To bridge these research gaps, we introduce CFALR (Collaborative Filtering-Augmented Large Language Model for Recommendation), a novel framework that synergizes collaborative filtering with large language models for personalized outfit recommendation. Specifically, CFALR describes user-outfit interactions in natural language and leverages LLMs to capture fashion semantics while employing CF-enhanced embeddings to bridge the semantic space and the collaborative interaction spaces. Our technical contributions include: (1) the first LLM-based architecture specifically designed for personalized outfit recommendation, (2) a CF-augmented generative mechanism that efficiently navigates the extensive combination space of outfit items, and (3) trainable projection layers that optimally integrate relational and content features. Experiments on Polyvore and IQON benchmarks demonstrate CFALR's superior performance over both traditional CF-based and LLM-based methods in personalized fill-in-the-blank and personalized outfit generation tasks.

Constant time programming patterns is the primary defense against timing attacks on cryptographic implementations, yet what "constant time" means varies across academia and industry. This work systematizes constant time models and their evolution, identifies a recurring gap between what models protect and what specifications assume, and distills an offensive methodology for discovering timing vulnerabilities that originate outside the cryptographic primitive boundary. Applying this methodology, we locate a specification-level vulnerability related to private key loading, and confirm the leak in both OpenSSL and BoringSSL. Counterintuitively, BoringSSL's per-observation signal is several orders of magnitude stronger than OpenSSL's, despite an explicitly stricter threat model.

Controlling the power output of a wind farm in order to track a target signal can be useful for the power grid frequency regulation. It can be achieved by dividing the target into individual setpoints, then followed by each turbines' controller. In this article, we are interested in finding power allocations that fairly spread the power reserves (i.e. unused fraction of available powers) among turbines, helping with robustness to uncertainties and changing wind conditions. In particular, we study the fairness properties of proportional dispatch, which is the most common power dispatching method. We show that due to the wake effects in a wind farm, proportional dispatch has to be applied iteratively to achieve fair distribution of power reserves. We study the convergence of this iterative process (referred to as IPD) to equalized reserves, and then illustrate it on simulated experiments, using steady-state and dynamic simulators. Numerical results show that IPD closely approaches max-min fairness, a related fairness objective, for a cheap computational price compared to black-box optimization. Finally, IPD is also shown to reduce the complexity of the problem of fair power dispatch combined with yaw wake steering optimization.

Generative Artificial Intelligence (GenAI), Large Language Models (LLMs), and emerging Agentic AI constitute the most disruptive transformation in the history of software engineering (SE), reshaping development processes, required competencies, professional roles, and the educational outcomes that universities must deliver. This paper presents a systematic review of 48 verified, influential peer-reviewed publications (2016--2026) drawn from leading venues in software engineering, machine learning, computing education, human--AI collaboration, and software productivity. Studies were discovered, screened, and analyzed through a four-agent research workflow (Literature Discovery, Scientometric Analysis, Curriculum Transformation, and Workforce Impact) and were verified against primary sources. We synthesize the evidence along nine themes and three trajectories -- practice, education, and workforce -- and report a scientometric inflection in which annual LLM-for-SE output grew roughly five-fold after late 2022. From this synthesis we contribute: (i) a conceptual framework for AI-native software engineering organized around \emph{intent}, \emph{collaboration}, and \emph{verification}; (ii) a nine-dimension competency model spanning specification, critical evaluation, agent orchestration, and metacognition; (iii) a four-phase university curriculum roadmap with AI-resilient assessment; (iv) faculty-development and workforce-transformation strategies; and (v) a prioritized agenda of eleven research gaps. The evidence base is internally contradictory on the magnitude and direction of productivity effects, underscoring that benefits are strongly context-dependent and that educating engineers for judgment, verification, and orchestration -- rather than code production alone -- is the central challenge of the AI-native era.

Conjoint analysis is a widely used preference measurement method in marketing research, political science, healthcare, and human-computer interaction. Despite broad adoption, researchers without access to commercial platforms face significant barriers, as existing tools are either expensive or lack end-to-end survey infrastructure. This paper presents an open-source, self-hosted web application for designing, deploying, and analysing conjoint surveys. Beyond conventional tabular stimuli, the platform uses generative AI to produce integrated stimuli formats: textual scenario descriptions generated by a large language model, and visual stimuli by a text-to-image model. A researcher-defined base prompt is parameterised with the conjoint profile, and optional LLM-facing level annotations enrich the generation. A structured setup wizard, AI-assisted attribute suggestion, and live data analysis lower the technical barriers for researchers new to conjoint methodology. A full export bundle including all stimuli, their generating prompts, and response data facilitates transparency and reproducibility. The platform is demonstrated through a proof-of-concept study on care robot preferences for ambient assisted living (AAL, N=55) using AI-generated visual stimuli. The paper discusses the role of AI assistance in conjoint design, arguing that theoretical grounding must remain the researcher's responsibility, and outlining how genAI-generated stimuli can broaden the methodological repertoire for HCI and related fields.

Binary search is deceptively simple in concept yet notoriously difficult to implement correctly. This paper presents a unified treatment of binary search: five core variants, six derived query functions, and four standard library implementations (BSD, glibc, Java, C++ STL), each with consistent notation, loop invariants, and analysis. We introduce bsearch_ultimate, a combined search that subsumes all variants in a single call. Every algorithm is provided as synchronized Python code, Dafny formal proof, and pseudocode. All implementations are validated by over 9,500 tests and 21 Dafny formal verifications; an additional six deliberately faulty implementations demonstrate common bug categories and Dafny's ability to detect them. We also provide memorable rules linking boundary choices to loop conditions and update formulas.