Related papers: MOVis: A Visual Analytics Tool for Surfacing Misse…

Predictive applications of machine learning often rely on small (sub 1 Bn parameter) specialized models tuned to particular domains or modalities. Such models often achieve excellent performance, but lack flexibility. LLMs and VLMs offer…

We study how to allocate a fixed supervised fine-tuning budget when three objectives must be balanced at once: multi-turn safety alignment, low over-refusal on benign boundary queries, and instruction following under verifiable constraints.…

This paper introduces cozy, a tool for analyzing and visualizing differences between two versions of a software binary. The primary use case for cozy is validating "micropatches": small binary or assembly-level patches inserted into…

The promise of multimodal models for real-world applications has inspired research in visualizing and understanding their internal mechanics with the end goal of empowering stakeholders to visualize model behavior, perform model debugging,…

We consider the task of semi-supervised video object segmentation (VOS). Our approach mitigates shortcomings in previous VOS work by addressing detail preservation and temporal consistency using visual warping. In contrast to prior work…

Metamorphic viruses engage different mutation techniques to escape from string signature based scanning. They try to change their code in new offspring so that the variants appear non-similar and have no common sequences of string as…

Code annotations is a widely used feature in Java systems to configure custom metadata on programming elements. Their increasing presence creates the need for approaches to assess and comprehend their usage and distribution. In this…

Zero-day vulnerabilities can be accidentally or maliciously placed in code and can remain in place for years. In this study, we address an aspect of their longevity by considering the likelihood that they will be discovered in the code…

Graphic visual content helps in promoting information communication and inspiration divergence. However, the interpretation of visual content currently relies mainly on humans' personal knowledge background, thereby affecting the quality…

A major bottleneck of interdisciplinary computer vision (CV) research is the lack of a framework that eases the reuse and abstraction of state-of-the-art CV models by CV and non-CV researchers alike. We present here BU-CVKit, a computer…

Vision Transformers (ViTs) are built on the assumption of treating image patches as ``visual tokens" and learn patch-to-patch attention. The patch embedding based tokenizer has a semantic gap with respect to its counterpart, the textual…

Collaborative autonomous driving with multiple vehicles usually requires the data fusion from multiple modalities. To ensure effective fusion, the data from each individual modality shall maintain a reasonably high quality. However, in…

In non-linear systems, where explicit analytic solutions usually can't be found, visualisation is a powerful approach which can give insights into the dynamical behaviour of models; it is also crucial for teaching this area of mathematics.…

We present MBIS (Multivariate Bayesian Image Segmentation tool), a clustering tool based on the mixture of multivariate normal distributions model. MBIS supports multi-channel bias field correction based on a B-spline model. A second…

One of the main challenges of debugging is to understand why the program fails for certain inputs but succeeds for others. This becomes especially difficult if the fault is caused by an interaction of multiple inputs. To debug such…

The rapidly developing deep learning (DL) techniques have been applied in software systems with various application scenarios. However, they could also pose new safety threats with potentially serious consequences, especially in…

Multi-view clustering can explore consistent information from different views to guide clustering. Most existing works focus on pursuing shallow consistency in the feature space and integrating the information of multiple views into a…

Understanding the functional (dis)-similarity of source code is significant for code modeling tasks such as software vulnerability and code clone detection. We present DISCO(DIS-similarity of COde), a novel self-supervised model focusing on…

Understanding the changes made by developers when they submit a pull request and/or perform a commit on a repository is a crucial activity in software maintenance and evolution. The common way to review changes relies on examining code…

Visual comparison is an important task in the analysis of multivariate graphs. However, comparison of topological features of a graph with respect to its data attributes for different portions of the data remains challenging because there…