arXiv subjects

Jaakko Peltonen

Publications and source records attributed to Jaakko Peltonen.

11 recordsLinked to original sources

arXiv2026

When Passing Tests Hides Vulnerabilities: An Empirical Study of Silent Failures in Agentic Systems

LLM-based agents for automated code repair have received significant attention in recent years from both research and software engineering practice perspectives. However, limited attention has been paid to patches that pass syntactic and functional verification but still retain or introduce security vulnerabilities. The aim of this research is to systematically identify and categorize such silent failures in LLM-based agentic code repair. We conducted an empirical study using 1,030 valid execution traces produced by seven agent frameworks with GPT-4o-mini across two security-focused datasets, SecurityEval and CVEfixes. Through three iterations of qualitative coding and manual verification, 170 confirmed silent failures were identified. The key results are: (i) Three main categories of silent failures were identified: Omission, Introduction, and Inadequacy. Omission accounts for 48.2% of the confirmed failures, Introduction for 30.6%, and Inadequacy for 21.2%. (ii) Ten fine-grained failure codes were classified under these three categories, showing how agents omit required security controls, apply incomplete defenses, or introduce new vulnerabilities during repair. (iii) Current test-passing evaluation and LLM-based reviewer roles were insufficient to expose or intercept these failures in the confirmed cases. (iv) Similar insecure solutions appeared across different frameworks, suggesting possible shared model-, prompt-, or task-level influences, while single-agent and multi-agent systems showed different failure profiles. The results of this study will assist researchers and practitioners in improving the evaluation of LLM-based agentic code repair and developing targeted verification methods that go beyond functional correctness and cover all generated artifacts.

cs.SE↗

arXiv2026

When Fireflies Cluster; Enhancing Automatic Clustering via Centroid-Guided Firefly Optimization

This work presents a novel variant of the Firefly Algorithm (FA) for data clustering, addressing limitations of traditional methods like K-Means that struggle with non-uniform cluster shapes, densities, and the need for pre-defining the number of clusters. The proposed algorithm introduces a centroid movement strategy and a multi-objective fitness function that balances compactness, separation, and a novel TSP-based navigation penalty. It automatically estimates the optimal number of clusters and dynamically adjusts cluster boundaries. Application to robotic sensor networks highlights its practical value, with experiments showing improved clustering quality and reduced intra-cluster path distances compared to K-Means. These results confirm the algorithm's robustness in complex spatial clustering tasks, with potential for future extensions to higher-dimensional and adaptive scenarios.

cs.AI↗

arXiv2025

Constrained Non-negative Matrix Factorization for Guided Topic Modeling of Minority Topics

Topic models often fail to capture low-prevalence, domain-critical themes, so-called minority topics, such as mental health themes in online comments. While some existing methods can incorporate domain knowledge, such as expected topical content, methods allowing guidance may require overly detailed expected topics, hindering the discovery of topic divisions and variation. We propose a topic modeling solution via a specially constrained NMF. We incorporate a seed word list characterizing minority content of interest, but we do not require experts to pre-specify their division across minority topics. Through prevalence constraints on minority topics and seed word content across topics, we learn distinct data-driven minority topics as well as majority topics. The constrained NMF is fitted via Karush-Kuhn-Tucker (KKT) conditions with multiplicative updates. We outperform several baselines on synthetic data in terms of topic purity, normalized mutual information, and also evaluate topic quality using Jensen-Shannon divergence (JSD). We conduct a case study on YouTube vlog comments, analyzing viewer discussion of mental health content; our model successfully identifies and reveals this domain-relevant minority content.

cs.LG↗

arXiv2023

Forecasting Emergency Department Crowding with Advanced Machine Learning Models and Multivariable Input

Emergency department (ED) crowding is a significant threat to patient safety and it has been repeatedly associated with increased mortality. Forecasting future service demand has the potential patient outcomes. Despite active research on the subject, several gaps remain: 1) proposed forecasting models have become outdated due to quick influx of advanced machine learning models (ML), 2) amount of multivariable input data has been limited and 3) discrete performance metrics have been rarely reported. In this study, we document the performance of a set of advanced ML models in forecasting ED occupancy 24 hours ahead. We use electronic health record data from a large, combined ED with an extensive set of explanatory variables, including the availability of beds in catchment area hospitals, traffic data from local observation stations, weather variables, etc. We show that N-BEATS and LightGBM outpeform benchmarks with 11 % and 9 % respective improvements and that DeepAR predicts next day crowding with an AUC of 0.76 (95 % CI 0.69-0.84). To the best of our knowledge, this is the first study to document the superiority of LightGBM and N-BEATS over statistical benchmarks in the context of ED forecasting.

cs.LG↗

arXiv2021

Edge-Path Bundling: A Less Ambiguous Edge Bundling Approach

Edge bundling techniques cluster edges with similar attributes (i.e. similarity in direction and proximity) together to reduce the visual clutter. All edge bundling techniques to date implicitly or explicitly cluster groups of individual edges, or parts of them, together based on these attributes. These clusters can result in ambiguous connections that do not exist in the data. Confluent drawings of networks do not have these ambiguities, but require the layout to be computed as part of the bundling process. We devise a new bundling method, Edge-Path bundling, to simplify edge clutter while greatly reducing ambiguities compared to previous bundling techniques. Edge-Path bundling takes a layout as input and clusters each edge along a weighted, shortest path to limit its deviation from a straight line. Edge-Path bundling does not incur independent edge ambiguities typically seen in all edge bundling methods, and the level of bundling can be tuned through shortest path distances, Euclidean distances, and combinations of the two. Also, directed edge bundling naturally emerges from the model. Through metric evaluations, we demonstrate the advantages of Edge-Path bundling over other techniques.

cs.GR↗

arXiv2019

Scalable Probabilistic Matrix Factorization with Graph-Based Priors

In matrix factorization, available graph side-information may not be well suited for the matrix completion problem, having edges that disagree with the latent-feature relations learnt from the incomplete data matrix. We show that removing these $\textit{contested}$ edges improves prediction accuracy and scalability. We identify the contested edges through a highly-efficient graphical lasso approximation. The identification and removal of contested edges adds no computational complexity to state-of-the-art graph-regularized matrix factorization, remaining linear with respect to the number of non-zeros. Computational load even decreases proportional to the number of edges removed. Formulating a probabilistic generative model and using expectation maximization to extend graph-regularised alternating least squares (GRALS) guarantees convergence. Rich simulated experiments illustrate the desired properties of the resulting algorithm. On real data experiments we demonstrate improved prediction accuracy with fewer graph edges (empirical evidence that graph side-information is often inaccurate). A 300 thousand dimensional graph with three million edges (Yahoo music side-information) can be analyzed in under ten minutes on a standard laptop computer demonstrating the efficiency of our graph update.

cs.LG↗

arXiv2019

Computing Stable Demers Cartograms

Cartograms are popular for visualizing numerical data for map regions. Maintaining correct adjacencies is a primary quality criterion for cartograms. When there are multiple data values per region (over time or different datasets) shown as animated or juxtaposed cartograms, preserving the viewer's mental-map in terms of stability between cartograms is another important criterion. We present a method to compute stable Demers cartograms, where each region is shown as a square and similar data yield similar cartograms. We enforce orthogonal separation constraints with linear programming, and measure quality in terms of keeping adjacent regions close (cartogram quality) and using similar positions for a region between the different data values (stability). Our method guarantees ability to connect most lost adjacencies with minimal leaders. Experiments show our method yields good quality and stability.

cs.CG↗

arXiv2016

Peacock Bundles: Bundle Coloring for Graphs with Globality-Locality Trade-off

Bundling of graph edges (node-to-node connections) is a common technique to enhance visibility of overall trends in the edge structure of a large graph layout, and a large variety of bundling algorithms have been proposed. However, with strong bundling, it becomes hard to identify origins and destinations of individual edges. We propose a solution: we optimize edge coloring to differentiate bundled edges. We quantify strength of bundling in a flexible pairwise fashion between edges, and among bundled edges, we quantify how dissimilar their colors should be by dissimilarity of their origins and destinations. We solve the resulting nonlinear optimization, which is also interpretable as a novel dimensionality reduction task. In large graphs the necessary compromise is whether to differentiate colors sharply between locally occurring strongly bundled edges ("local bundles"), or also between the weakly bundled edges occurring globally over the graph ("global bundles"); we allow a user-set global-local tradeoff. We call the technique "peacock bundles". Experiments show the coloring clearly enhances comprehensibility of graph layouts with edge bundling.

cs.CG↗

arXiv2016

An Information Retrieval Approach to Finding Dependent Subspaces of Multiple Views

Finding relationships between multiple views of data is essential both for exploratory analysis and as pre-processing for predictive tasks. A prominent approach is to apply variants of Canonical Correlation Analysis (CCA), a classical method seeking correlated components between views. The basic CCA is restricted to maximizing a simple dependency criterion, correlation, measured directly between data coordinates. We introduce a new method that finds dependent subspaces of views directly optimized for the data analysis task of \textit{neighbor retrieval between multiple views}. We optimize mappings for each view such as linear transformations to maximize cross-view similarity between neighborhoods of data samples. The criterion arises directly from the well-defined retrieval task, detects nonlinear and local similarities, is able to measure dependency of data relationships rather than only individual data coordinates, and is related to well understood measures of information retrieval quality. In experiments we show the proposed method outperforms alternatives in preserving cross-view neighborhood similarities, and yields insights into local dependencies between multiple views.

stat.ML↗

arXiv2015

Genome-wide modelling of transcription kinetics reveals patterns of RNA production delays

Genes with similar transcriptional activation kinetics can display very different temporal mRNA profiles due to differences in transcription time, degradation rate and RNA processing kinetics. Recent studies have shown that a splicing-associated RNA production delay can be significant. We introduce a joint model of transcriptional activation and mRNA accumulation which can be used for inference of transcription rate, RNA production delay and degradation rate given genome-wide data from high-throughput sequencing time course experiments. We combine a mechanistic differential equation model with a non-parametric statistical modelling approach allowing us to capture a broad range of activation kinetics, and use Bayesian parameter estimation to quantify the uncertainty in the estimates of the kinetic parameters. We apply the model to data from estrogen receptor (ER-α) activation in the MCF-7 breast cancer cell line. We use RNA polymerase II (pol-II) ChIP-Seq time course data to characterise transcriptional activation and mRNA-Seq time course data to quantify mature transcripts. We find that 11% of genes with a good signal in the data display a delay of more than 20 minutes between completing transcription and mature mRNA production. The genes displaying these long delays are significantly more likely to be short. We also find a statistical association between high delay and late intron retention in pre-mRNA data, indicating significant splicing-associated production delays in many genes.

q-bio.GN↗

arXiv2014

Toward computational cumulative biology by combining models of biological datasets

A main challenge of data-driven sciences is how to make maximal use of the progressively expanding databases of experimental datasets in order to keep research cumulative. We introduce the idea of a modeling-based dataset retrieval engine designed for relating a researcher's experimental dataset to earlier work in the field. The search is (i) data-driven to enable new findings, going beyond the state of the art of keyword searches in annotations, (ii) modeling-driven, to both include biological knowledge and insights learned from data, and (iii) scalable, as it is accomplished without building one unified grand model of all data. Assuming each dataset has been modeled beforehand, by the researchers or by database managers, we apply a rapidly computable and optimizable combination model to decompose a new dataset into contributions from earlier relevant models. By using the data-driven decomposition we identify a network of interrelated datasets from a large annotated human gene expression atlas. While tissue type and disease were major driving forces for determining relevant datasets, the found relationships were richer and the model-based search was more accurate than keyword search; it moreover recovered biologically meaningful relationships that are not straightforwardly visible from annotations, for instance, between cells in different developmental stages such as thymocytes and T-cells. Data-driven links and citations matched to a large extent; the data-driven links even uncovered corrections to the publication data, as two of the most linked datasets were not highly cited and turned out to have wrong publication entries in the database.

q-bio.QM↗