Natural Language to SQL (NL2SQL) enables natural language access to structured data, but LLM-based methods can be inefficient for real-time use and repetitive query patterns. We present AutoRuleSQL, a hybrid system that combines template-based fast paths with LLM fallback and offline bootstrapping. Empirical results show that it reduces latency by over 12.6% and improves execution accuracy by up to 4.0%, when combined with existing NL2SQL methods.

Reinforcement learning (RL) has emerged as a promising tool for combinatorial optimization (CO) problems due to its ability to learn fast, effective, and generalizable solutions. Nonetheless, existing works mostly focus on one-shot deterministic CO, while sequential stochastic CO (SSCO) has rarely been studied despite its broad applications such as adaptive influence maximization (IM) and infectious disease intervention. In this paper, we study the SSCO problem where we first decide the budget (e.g., number of seed nodes in adaptive IM) allocation for all time steps, and then select a set of nodes for each time step. The few existing studies on SSCO simplify the problems by assuming a uniformly distributed budget allocation over the time horizon, yielding suboptimal solutions. We propose a generic hierarchical RL (HRL) framework called wake-sleep option (WS-option), a two-layer option-based framework that simultaneously decides adaptive budget allocation on the higher layer and node selection on the lower layer. WS-option starts with a coherent formulation of the two-layer Markov decision processes (MDPs), capturing the interdependencies between the two layers of decisions. Building on this, WS-option employs several innovative designs to balance the model's training stability and computational efficiency, preventing the vicious cyclic interference issue between the two layers. Empirical results show that WS-option exhibits significantly improved effectiveness and generalizability compared to traditional methods. Moreover, the learned model can be generalized to larger graphs, which significantly reduces the overhead of computational resources.

Path planning is a critical challenge for Autonomous Underwater Vehicles (AUVs) due to complex underwater environments, including ocean currents, dynamic obstacles, and limited sensing capabilities. The lack of a standardized benchmarking framework has hindered direct comparisons between algorithms, slowing progress in the field. To address this, we introduce an open-source benchmarking platform for underwater AUV path planning, designed to provide a unified evaluation environment, automated performance assessment, and reproducible experiments. Built on the HoloOcean simulation platform, our benchmark incorporates realistic underwater dynamics, such as ocean currents, static and dynamic obstacles, and sensor models. It supports a range of path planning tasks, from basic obstacle avoidance to complex scenarios with current disturbances. The platform is compatible with classical algorithms (e.g., A*, RRT), evolutionary methods (e.g., GA, ACO), and deep reinforcement learning (e.g., Soft Actor-Critic, SAC). We define key evaluation metrics, including path efficiency (length, smoothness, energy consumption), task success rate, collision rate, and computational cost. Automated tools enable systematic algorithm comparisons across scenarios, generating standardized performance results and visualizations. This open-source, extensible framework aims to advance underwater path planning research by enabling fair comparisons and guiding future algorithm development. It provides a scalable foundation for evaluating AUV path planning methods under simulated real-world conditions, fostering innovation in the field. All source code and experimental configurations will be available on the GitHub Page: https://github.com/IoET-y/UP-bench.

Abstract Android malware is the major cyber threat to the popular Android platform which may influence millions of end users. To battle against the Android malware, a large number of machine learning methods either based on 1) traditional feature extraction using static and dynamic analysis, or 2) recently proposed image representations, have been developed, and have achieved promising results. However, the vast majority of the existing work rely on a large number of labeled samples which are unfortunately not available for the newly reported Android malware families. This poses a critical challenge to detect such few-shot Android malware families. In this paper, we propose a novel few-shot learning approach based on the image representation of an Android application to solve the problem. With an application file converted into an image representation, we preserve all the source code information. We then utilize self-supervised learning to obtain the pre-trained backbone from the unlabeled auxiliary data and employ a metric-based few-shot learning method for Android malware classification. Considering the impact of irrelevant information across samples on the family classification, we employ a multi-cropping strategy to capture family label-related information in the images. Extensive experimental results on the popular CICInvesAndMal2019 dataset confirm the effectiveness of our approach in detecting few-shot Android malware families. We achieve at least 3.16% and 3.7% improvement on 5-way 1-shot and 5-way 5-shot scenarios respectively comparing to state-of-the-art baselines.

Months may pass before the victim of IP theft even knows they have been compromised. During this time, the attacker can exfiltrate large amounts of data. Recent work has proposed the idea of injecting a set of believable fake versions of a real document into a network so that the attacker has to expend time and effort to identify the real document from a sea of similar documents. In this paper, we consider the problem of an attacker who is smart and breaks a technical document down into small, bit-sized “units” and inspects them one by one so as to defeat the fake document defense. If a unit in a document is determined to be fake, the adversary does not need to look further at the same document. He can also immediately identify as fake, any other document that contains the same unit. In this paper, we consider the problem of a smart attacker using this strategy. Our proposed defensive algorithm, called${\sf GAIT}$, is shown to be successful in mitigating such attacks.${\sf GAIT}$can work in conjunction with any NLP-based generative method to create fake technical documents.

Past research has shown that over 8 months may elapse between the time when a network is compromised and the time the attack is discovered. During this long gap, attackers can steal valuable intellectual property from the victim. The recent FORGE system [8] has suggested that automatically generating fake—but believable—versions of documents can delay the attacker, cost him money, and increase his uncertainty. However, in order to generate fakes, FORGE only modifies the textual component of the document in question. But in the real world, documents consist of many non-textual components such as charts, equations, formulas, diagrams, and tables. We propose the concept of a Probabilistic Logic Graph (PLG) and show that PLGs provide a single, unified framework within which the different parts of a document can be expressed. We then define the problem of generating, for a given PLG representation of a document, a set of fake yet highly believable PLGs (i.e., documents), so that an attacker looking at them (both the original and the fake ones) cannot easily identify the original document. We show that the problem of generating fake PLGs is intractable—but we propose an approximation algorithm that solves it efficiently. We evaluate the use of PLGs over a corpus of patents and show that our fakes can effectively deceive an adversary.

According to Symantec, the average gap from the time a company is compromised by a zero-day attack to the time the vulnerability is discovered is 312 days. This leaves an adversary with a lot of time to exfiltrate corporate IP. Recent work has suggested automatically generating multiple fake versions of a document to impose costs on the attacker who needs to correctly identify the original document from a set of mostly fake documents. But in the real world, documents contain many diverse components. In this article, we focus on technical documents that often contain equations. We present${\sf FEE}$(Fake Equation Engine), a framework to generate fake equations in such documents.${\sf FEE}$tries to preserve multiple aspects of a given equation when generating a fake. Moreover,${\sf FEE}$is very general and applies to diverse equational forms including polynomial equations, differential equations, transcendental equations, and more.${\sf FEE}$iteratively solves a complex, changing optimization problem inside it. We also present${\sf FEE-FAST}$, a fast approximate algorithm to solve the optimization problem within${\sf FEE}$. Using a panel of human subjects, we show that${\sf FEE}$achieves a high rate in deceiving sophisticated subjects.

Although most game theory models assume that payoff matrices are provided as input, getting payoff matrices in strategic games (e.g., corporate negotiations and counter-terrorism operations) has proven difficult. To tackle this challenge, we propose a payoff inference engine (PIE) that finds payoffs assuming that players in a game follow a myopic best response or a regret minimization heuristic. This assumption yields a set of constraints (possibly nonlinear) on the payoffs with a multiplicity of solutions. PIE finds payoffs by considering solutions of these constraints and their variants via three heuristics. First, we approximately compute a centroid of the resulting polytope of the constraints. Second, we use a soft constraint approach that allows violation of constraints by penalizing violations in the objective function. Third, we develop a novel approach to payoff inference based on support vector machines (SVMs). Unlike past work on payoff inference, PIE has the following advantages. PIE supports reasoning about multiplayer games, not just one or two players, it can use short histories, not long ones which may not be available in many real-world situations, it does not require all players to be fully rational, and it is one to two orders of magnitude more scalable than past work. We run experiments on a synthetic data set where we generate payoff functions for the players and see how well our algorithms can learn them, a real-world coarse-grained counter-terrorism data set about a set of different terrorist groups, and a real-world fine-grained data set about a specific terrorist group. As the ground truth about payoffs for the terrorist groups cannot be tested directly, we test PIE by using the payoffs to make predictions about the actions of the groups and corresponding governments (even though this is not the purpose of this article). We show that compared with recent work on payoff inference, PIE has both higher accuracy and much shorter runtime.

As the most widely used OS on earth, Android is heavily targeted by malicious hackers. Though much work has been done on detecting Android malware, hackers are becoming increasingly adept at evading ML classifiers. We develop$\textsf {FARM}$, a Feature transformation basedAndRoidMalware detector.$\textsf {FARM}$takes well-known features for Android malware detection and introduces three new types of feature transformations that transform these features irreversibly into a new feature domain. We first test$\textsf {FARM}$on 6 Android classification problems separating goodware and “other malware” from 3 classes of malware: rooting malware, spyware, and banking trojans. We show that$\textsf {FARM}$beats standard baselines when no attacks occur. Though we cannot guess all possible attacks that an adversary might use, we propose three realistic attacks on$\textsf {FARM}$and show that$\textsf {FARM}$is very robust to these attacks in all classification problems. Additionally,$\textsf {FARM}$has automatically identified two malware samples which were not previously classified as rooting malware by any of the 61 anti-viruses on VirusTotal. These samples were reported to Google’s Android Security Team who subsequently confirmed our findings.

It has been an open challenge for self-interested agents to make optimal sequential decisions in complex multiagent systems, where agents might achieve higher utility via collaboration. The Microsoft Malmo Collaborative AI Challenge (MCAC), which is designed to encourage research relating to various problems in Collaborative AI, takes the form of a Minecraft mini-game where players might work together to catch a pig or deviate from cooperation, for pursuing high scores to win the challenge. Various characteristics, such as complex interactions among agents, uncertainties, sequential decision making and limited learning trials all make it extremely challenging to find effective strategies. We present HogRider---the champion agent of MCAC in 2017 out of 81 teams from 26 countries. One key innovation of HogRider is a generalized agent type hypothesis framework to identify the behavior model of the other agents, which is demonstrated to be robust to observation uncertainty. On top of that, a second key innovation is a novel Q-learning approach to learn effective policies against each type of the collaborating agents. Various ideas are proposed to adapt traditional Q-learning to handle complexities in the challenge, including state-action abstraction to reduce problem scale, a warm start approach using human reasoning for addressing limited learning trials, and an active greedy strategy to balance exploitation-exploration. Challenge results show that HogRider outperforms all the other teams by a significant edge, in terms of both optimality and stability.

To reduce the air pollution and improve the energy efficiency, many countries and cities (e.g., Singapore) are on the way of introducing electric vehicles (EVs) to replace the vehicles serving in current traffic system. Effective placement of charging stations is essential for the rapid development of EVs, because it is necessary for providing convenience for EVs and ensuring the efficiency of the traffic network. However, existing works mostly concentrate on the mileage anxiety from EV users but ignore their strategic and competitive charging behaviors. To capture the competitive and strategic charging behaviors of the EV users, we consider that an EV user’s charging cost, which is dependent on other EV users’ choices, consists of the travel cost to access the charging station and the queuing cost in charging stations. First, we formulate the Charging Station Placement Problem (CSPP) as a bilevel optimization problem. Then, by exploiting the equilibrium of the EV charging game, we convert the bilevel optimization problem to a single-level one, following which we analyze the properties of CSPP and propose an algorithm Optimizing eleCtric vEhicle chArging statioN (OCEAN) to compute the optimal allocation of charging stations. Due to OCEAN’s scalability issue, we furthermore present a heuristic algorithm OCEAN with Continuous variables to deal with large-scale real-world problems. Finally, we demonstrate and discuss the results of the extensive experiments we did. It is shown that our approach outperform baseline methods significantly.