arXiv subjects

Lars Kulik

Publications and source records attributed to Lars Kulik.

8 recordsLinked to original sources

arXiv2026

ARBITER: Reasoning Trajectory Basins and Majority Vote Failures in Test-Time Sampling

When language models use test-time sampling, they generate multiple reasoning trajectories and select an answer by majority vote. We show that these trajectories are not independent: for a given question, they concentrate into a small number of clusters, or reasoning basins, each defined by a normalized final answer and the solutions that reach it. A majority vote therefore selects the most stable basin rather than the most accurate one, which creates wrong-majority failures where the correct answer is present but outvoted. We introduce ARBITER, a model-agnostic approach that models interactions between basins using only the base model's own sampled outputs, hidden states, and derived evidence. Most direct correction strategies fail; ARBITER instead uses conservative additive evidence on top of consensus. In its simplest parameter-free form, ARBITER-{\Delta} adds same-model evidence to the majority prior, while ARBITER-Enc augments this with bounded residual signals from hidden states over complete solutions. On GSM8K with Qwen3-4B, consensus over K=24 samples achieves around the mid-94% range, while a same-pool top-2 oracle reaches around the mid-96% range. ARBITER recovers a subset of these cases using zero external information. Across three model families and three math benchmarks, it yields consistent gains with no net-negative cases; for example, on Llama-3.1-8B MMLU-HS-Math, it improves accuracy from the mid-78% range to the mid-82% range, recovering about 22% of the available oracle headroom, indicating that this headroom can be partially recovered from the sample pool itself.

cs.LG↗

arXiv2023

Efficient Cost Modeling of Space-filling Curves

A space-filling curve (SFC) maps points in a multi-dimensional space to one-dimensional points by discretizing the multi-dimensional space into cells and imposing a linear order on the cells. This way, an SFC enables the indexing of multi-dimensional data using a one-dimensional index such as a B+-tree. Choosing an appropriate SFC is crucial, as different SFCs have different effects on query performance. Currently, there are two primary strategies: 1) deterministic schemes, which are computationally efficient but often yield suboptimal query performance, and 2) dynamic schemes, which consider a broad range of candidate SFCs based on cost functions but incur significant computational overhead. Despite these strategies, existing methods cannot efficiently measure the effectiveness of SFCs under heavy query workloads and numerous SFC options. To address this problem, we propose means of constant-time cost estimations that can enhance existing SFC selection algorithms, enabling them to learn more effective SFCs. Additionally, we propose an SFC learning method that leverages reinforcement learning and our cost estimation to choose an SFC pattern efficiently. Experimental studies offer evidence of the effectiveness and efficiency of the proposed means of cost estimation and SFC learning.

cs.DB↗

arXiv2021

Fast, Accurate and Interpretable Time Series Classification Through Randomization

Time series classification (TSC) aims to predict the class label of a given time series, which is critical to a rich set of application areas such as economics and medicine. State-of-the-art TSC methods have mostly focused on classification accuracy and efficiency, without considering the interpretability of their classifications, which is an important property required by modern applications such as appliance modeling and legislation such as the European General Data Protection Regulation. To address this gap, we propose a novel TSC method - the Randomized-Supervised Time Series Forest (r-STSF). r-STSF is highly efficient, achieves state-of-the-art classification accuracy and enables interpretability. r-STSF takes an efficient interval-based approach to classify time series according to aggregate values of discriminatory sub-series (intervals). To achieve state-of-the-art accuracy, r-STSF builds an ensemble of randomized trees using the discriminatory sub-series. It uses four time series representations, nine aggregation functions and a supervised binary-inspired search combined with a feature ranking metric to identify highly discriminatory sub-series. The discriminatory sub-series enable interpretable classifications. Experiments on extensive datasets show that r-STSF achieves state-of-the-art accuracy while being orders of magnitude faster than most existing TSC methods. It is the only classifier from the state-of-the-art group that enables interpretability. Our findings also highlight that r-STSF is the best TSC method when classifying complex time series datasets.

cs.LG↗

arXiv2021

A Lazy Approach for Efficient Index Learning

Learned indices using neural networks have been shown to outperform traditional indices such as B-trees in both query time and memory. However, learning the distribution of a large dataset can be expensive, and updating learned indices is difficult, thus hindering their usage in practical applications. In this paper, we address the efficiency and update issues of learned indices through agile model reuse. We pre-train learned indices over a set of synthetic (rather than real) datasets and propose a novel approach to reuse these pre-trained models for a new (real) dataset. The synthetic datasets are created to cover a large range of different distributions. Given a new dataset DT, we select the learned index of a synthetic dataset similar to DT, to index DT. We show a bound over the indexing error when a pre-trained index is selected. We further show how our techniques can handle data updates and bound the resultant indexing errors. Experimental results on synthetic and real datasets confirm the effectiveness and efficiency of our proposed lazy (model reuse) approach.

cs.DB↗

arXiv2019

GeoPrune: Efficiently Finding Shareable Vehicles Based on Geometric Properties

On-demand ride-sharing is rapidly growing.Matching trip requests to vehicles efficiently is critical for the service quality of ride-sharing. To match trip requests with vehicles, a prune-and-select scheme is commonly used. The pruning stage identifies feasible vehicles that can satisfy the trip constraints (e.g., trip time). The selection stage selects the optimal one(s) from the feasible vehicles. The pruning stage is crucial to reduce the complexity of the selection stage and to achieve efficient matching. We propose an effective and efficient pruning algorithm called GeoPrune. GeoPrune represents the time constraints of trip requests using circles and ellipses, which can be computed and updated efficiently. Experiments on real-world datasets show that GeoPrune reduces the number of vehicle candidates in nearly all cases by an order of magnitude and the update cost by two to three orders of magnitude compared to the state-of-the-art.

cs.DB↗

arXiv2019

A Differentially Private Algorithm for Range Queries on Trajectories

We propose a novel algorithm to ensure $ε$-differential privacy for answering range queries on trajectory data. In order to guarantee privacy, differential privacy mechanisms add noise to either data or query, thus introducing errors to queries made and potentially decreasing the utility of information. In contrast to the state-of-the-art, our method achieves significantly lower error as it is the first data- and query-aware approach for such queries. The key challenge for answering range queries on trajectory data privately is to ensure an accurate count. Simply representing a trajectory as a set instead of \emph{sequence} of points will generally lead to highly inaccurate query answers as it ignores the sequential dependency of location points in trajectories, i.e., will violate the consistency of trajectory data. Furthermore, trajectories are generally unevenly distributed across a city and adding noise uniformly will generally lead to a poor utility. To achieve differential privacy, our algorithm adaptively adds noise to the input data according to the given query set. It first privately partitions the data space into uniform regions and computes the traffic density of each region. The regions and their densities, in addition to the given query set, are then used to estimate the distribution of trajectories over the queried space, which ensures high accuracy for the given query set. We show the accuracy and efficiency of our algorithm using extensive empirical evaluations on real and synthetic data sets.

cs.DB↗

arXiv2011

Privacy Preserving Moving KNN Queries

We present a novel approach that protects trajectory privacy of users who access location-based services through a moving k nearest neighbor (MkNN) query. An MkNN query continuously returns the k nearest data objects for a moving user (query point). Simply updating a user's imprecise location such as a region instead of the exact position to a location-based service provider (LSP) cannot ensure privacy of the user for an MkNN query: continuous disclosure of regions enables the LSP to follow a user's trajectory. We identify the problem of trajectory privacy that arises from the overlap of consecutive regions while requesting an MkNN query and provide the first solution to this problem. Our approach allows a user to specify the confidence level that represents a bound of how much more the user may need to travel than the actual kth nearest data object. By hiding a user's required confidence level and the required number of nearest data objects from an LSP, we develop a technique to prevent the LSP from tracking the user's trajectory for MkNN queries. We propose an efficient algorithm for the LSP to find k nearest data objects for a region with a user's specified confidence level, which is an essential component to evaluate an MkNN query in a privacy preserving manner; this algorithm is at least two times faster than the state-of-the-art algorithm. Extensive experimental studies validate the effectiveness of our trajectory privacy protection technique and the efficiency of our algorithm.

cs.DB↗

arXiv2009

Decentralized Traffic Management Strategies for Sensor-Enabled Cars

Traffic Congestions and accidents are major concerns in today's transportation systems. This thesis investigates how to optimize traffic flow on highways, in particular for merging situations such as intersections where a ramp leads onto the highway. In our work, cars are equipped with sensors that can detect distance to neighboring cars, and communicate their velocity and acceleration readings with one another. Sensor-enabled cars can locally exchange sensed information about the traffic and adapt their behavior much earlier than regular cars. We propose proactive algorithms for merging different streams of sensor-enabled cars into a single stream. A proactive merging algorithm decouples the decision point from the actual merging point. Sensor-enabled cars allow us to decide where and when a car merges before it arrives at the actual merging point. This leads to a significant improvement in traffic flow as velocities can be adjusted appropriately. We compare proactive merging algorithms against the conventional priority-based merging algorithm in a controlled simulation environment. Experiment results show that proactive merging algorithms outperform the priority-based merging algorithm in terms of flow and delay.

cs.DC↗