arXiv subjects

Siddharth Gupta

Publications and source records attributed to Siddharth Gupta.

At least 19 recordsLinked to original sources

arXiv2026

Cross-Model Agreement as a Deployment-Time Reliability Signal for Automatic Polyp Segmentation

In real-time colonoscopy, ground-truth annotations are unavailable at inference, so polyp segmentation models can fail silently. We propose Referee-Based Quality Estimation (RBQE), a reference-free framework measuring agreement between a primary segmentation model and an independently trained referee on the same image. RBQE is evaluated on a standardized 1,223-image external benchmark drawn from four public datasets, using four referee configurations chosen to separate two design axes: referee independence and architectural diversity. Using a common Agreement Dice descriptor, a same-architecture referee differing from the primary model only in random initialization already yields a useful reliability signal (ROC-AUC = 0.923), showing that independent training alone is sufficient. Cross-architecture referees improve further: SegFormer-B0 achieves the strongest performance (ROC-AUC = 0.960), significantly outperforming the same-architecture control and UNet++, and exceeding a representative Test-Time Augmentation baseline by 0.055 ROC-AUC under an identical protocol, whereas a prompt-coupled MedSAM referee underperforms despite maximal architectural diversity. Because empty-mask agreement is trivially separable, we also report a restricted evaluation excluding such cases: ROC-AUC falls to 0.876 (SegFormer-B0, 1,046 images) and 0.783 (same-architecture control, 975 images), yet RBQE's margin over both baselines widens on this identical subset. RBQE additionally increases the mean Dice of retained predictions as low-agreement cases are progressively rejected, supporting selective prediction, and requires only one additional deterministic referee forward pass at inference. Our study therefore supports cross-model agreement as a practical, interpretable reliability framework for automated polyp segmentation.

cs.CV↗

arXiv2026

Beyond Masks: The Case for Medical Image Parsing

Medical imaging research has spent a decade getting very good at one thing: producing per-voxel masks. Masks tell us size, volume, and location, and a decade of clinical infrastructure rests on those outputs. Yet the report a radiologist writes contains almost nothing a mask can express. We argue that medical imaging research should adopt medical image parsing as its central output: a structured representation in which entities, attributes, and relationships are emitted together and mutually consistent. Entities are the named structures and findings, present or absent. Attributes describe those entities, capturing things like margin regularity, enhancement pattern, or severity grade. Relationships connect them, naming where one structure sits relative to another, what abuts what, and what has changed since the prior scan. A good parse satisfies three properties, in order: (1) decision (the parse names the right things in the current image), (2) reconstruction (its content is rich enough to regenerate that image), and (3) prediction (its content is rich enough to forecast how the patient state will evolve). Quantitative measurements are derived from this content; they are not predicted alongside it. To test how close the field is to producing such an output, we audit eleven representative systems against the three parsing primitives plus closure. None emits a well-formed parse. Entities are largely solved. Attributes, relationships, and closure remain near-empty. The path forward is not a new architecture. It is a commitment to a richer output, and to training signals that reward it. Segmentation taught models to measure. Parsing asks them to explain.

cs.CV↗

arXiv2026

Human-AI Coordination Zones: A Framework for Designing Human-in-the-Loop Experiences with Agentic AI

As generative and agentic AI becomes embedded in everyday products, practitioners face a persistent challenge: how to design human-AI coordination -- the ongoing mutual adjustment between users and AI systems as mediate through interfaces-that supports usability, trust, and safety. Existing resources offer high-level principles ("be transparent," "maintain user control") or low-level UI patterns, but there is a lack of mid-level design knowledge bridging the two. Through landscape and artifact analysis of 60 commercial AI applications, we introduce a framework defining human-AI coordination as the interplay of three dimensions: salience (how prominently AI is presented), involvement (what users can do to engage AI), and activity (what AI actually does). We contribute mid-level tools including coordination zones (done-for-me, done-under-me, done-with-me, done-without-me), an input taxonomy (prompted, sparked, inferred, layered), coordination curves for mapping user journeys, and design patterns demonstrating the generative capacity of the framework. The framework can be applied generatively to design experiences, analytically to evaluate existing ones, and communicatively to articulate ideas across stakeholders.

cs.HC↗

arXiv2026

On the Complexity of Extending Storylines

Storyline layouts visualize temporal interactions by drawing each character as an x-monotone curve and enforcing that the participants of every meeting form a contiguous vertical group. We study a drawing extension variant in which a layout of a sub-storyline is fixed and has to be extended by inserting $k$ missing characters while adhering to all meeting constraints. We consider two optimization objectives: minimizing the number of additional crossings introduced to complete the storyline, and minimizing the maximum number of crossings incurred by any single character. For both variants, we analyze the parameterized complexity with respect to the natural parameter $k$ representing the size of the missing information, as well as additional structural parameters such as the number of characters $\sigma$ per time instant and the number of meetings $\mu$ involving missing characters. We contribute a broad collection of results, most of them tight, drawing a nearly complete picture of the complexity landscape of these extension problems.

cs.CG↗

arXiv2025

Hypergraphs as Metro Maps: Drawing Paths with Few Bends in Trees, Cacti, and Plane 4-Graphs

A hypergraph consists of a set of vertices and a set of subsets of vertices, called hyperedges. In the metro map metaphor, each hyperedge is represented by a path (the metro line) and the union of all these paths is the support graph (metro network) of the hypergraph. Formally speaking, a path-based support is a graph together with a set of paths. We consider the problem of constructing drawings of path-based supports that (i) minimize the sum of the number of bends on all paths, (ii) minimize the maximum number of bends on any path, or (iii) maximize the number of 0-bend paths, then the number of 1-bend paths, etc. We concentrate on straight-line drawings of path-based tree and cactus supports as well as orthogonal drawings of path-based plane supports with maximum degree 4.

cs.CG↗

arXiv2025

Closed Neighborhood Balanced k-Coloring of Graphs

For a simple graph G = (V, E) and a positive integer k greater than or equal to 2, a coloring of vertices of G using exactly k colors such that every vertex has an equal number of vertices of each color in its closed neighborhood is called closed neighborhood-balanced k-coloring, and the graph which admits such a coloring is called closed neighborhood balanced k-colored graph. We derive some necessary/sufficient conditions for a graph to admit a closed neighborhood balanced k-coloring and discuss various graph operations involving such graphs. Furthermore, we prove that there is no forbidden subgraph characterization for the class of closed neighborhood-balanced k-colorable graphs.

math.CO↗

arXiv2025

YOLO-LAN: Precise Polyp Detection via Optimized Loss, Augmentations and Negatives

Colorectal cancer (CRC), a lethal disease, begins with the growth of abnormal mucosal cell proliferation called polyps in the inner wall of the colon. When left undetected, polyps can become malignant tumors. Colonoscopy is the standard procedure for detecting polyps, as it enables direct visualization and removal of suspicious lesions. Manual detection by colonoscopy can be inconsistent and is subject to oversight. Therefore, object detection based on deep learning offers a better solution for a more accurate and real-time diagnosis during colonoscopy. In this work, we propose YOLO-LAN, a YOLO-based polyp detection pipeline, trained using M2IoU loss, versatile data augmentations and negative data to replicate real clinical situations. Our pipeline outperformed existing methods for the Kvasir-seg and BKAI-IGH NeoPolyp datasets, achieving mAP$_{50}$ of 0.9619, mAP$_{50:95}$ of 0.8599 with YOLOv12 and mAP$_{50}$ of 0.9540, mAP$_{50:95}$ of 0.8487 with YOLOv8 on the Kvasir-seg dataset. The significant increase is achieved in mAP$_{50:95}$ score, showing the precision of polyp detection. We show robustness based on polyp size and precise location detection, making it clinically relevant in AI-assisted colorectal screening.

cs.CV↗

arXiv2025

Neighborhood Balanced k-Coloring of Graphs

For a simple graph G = (V, E) and a positive integer k greater than or equal to 2, a coloring of vertices of G using exactly k colors such that each vertex has an equal number of neighbors of each color is called neighborhood-balanced k-coloring, and the graph is called a neighborhood-balanced k-colored graph. This generalizes the notion of neighborhood balanced coloring of graphs introduced by Bryan Freyberg and Alison Marr (Graphs and Combinatorics, 2024). We derive some necessary/sufficient conditions for a graph to admit a neighborhood-balanced k-coloring and discuss several graph classes that admit such colorings. We also show that the problem of determining whether a given graph has such a coloring is NP-complete. Furthermore, we prove that there is no forbidden subgraph characterization for the class of neighborhood-balanced k-colorable graphs.

math.CO↗

arXiv2025

Efficient Distributed Algorithms for Shape Reduction via Reconfigurable Circuits

Autonomous reconfiguration of agent-based systems is a key challenge in the study of programmable matter, distributed robotics, and molecular self-assembly. While substantial prior work has focused on size-preserving transformations, much less is known about size-changing transformations. Such transformations find application in natural processes, active self-assembly, and dynamical systems, where structures may evolve through the addition or removal of components controlled by local rules. In this paper, we study efficient distributed algorithms for transforming 2D geometric configurations of simple agents, called shapes, using only local size-changing operations. A novelty of our approach is the use of reconfigurable circuits as the underlying communication model, a recently proposed model enabling instant node-to-node communication via primitive signals. Unlike previous work, we integrate collision avoidance as a core responsibility of the distributed algorithm. We consider two graph update models: connectivity and adjacency. Let $n$ denote the number of agents and $k$ the number of turning points in the initial shape. In the connectivity model, we show that any tree-shaped configuration can be reduced to a single agent using only shrinking operations in $O(k \log n)$ rounds w.h.p., and to its incompressible form in $O(\log n)$ rounds w.h.p. given prior knowledge of the incompressible nodes, or in $O(k \log n)$ rounds otherwise. When both shrinking and growth operations are available, we give an algorithm that transforms any tree to a topologically equivalent one in $O(k \log n + \log^2 n)$ rounds w.h.p. On the negative side, we show that one cannot hope for $o(\log^2 n)$-round transformations for all shapes of $\Theta(\log n)$ turning points. In the adjacency model, we show that any connected shape can reduce itself to a single node using only shrinking in $O(\log n)$ rounds w.h.p.

cs.DC↗

arXiv2024

Exact Algorithms for Clustered Planarity with Linear Saturators

We study Clustered Planarity with Linear Saturators, which is the problem of augmenting an $n$-vertex planar graph whose vertices are partitioned into independent sets (called clusters) with paths - one for each cluster - that connect all the vertices in each cluster while maintaining planarity. We show that the problem can be solved in time $2^{O(n)}$ for both the variable and fixed embedding case. Moreover, we show that it can be solved in subexponential time $2^{O(\sqrt{n}\log n)}$ in the fixed embedding case if additionally the input graph is connected. The latter time complexity is tight under the Exponential-Time Hypothesis. We also show that $n$ can be replaced with the vertex cover number of the input graph by providing a linear (resp. polynomial) kernel for the variable-embedding (resp. fixed-embedding) case; these results contrast the NP-hardness of the problem on graphs of bounded treewidth (and even on trees). Finally, we complement known lower bounds for the problem by showing that Clustered Planarity with Linear Saturators is NP-hard even when the number of clusters is at most $3$, thus excluding the algorithmic use of the number of clusters as a parameter.

cs.DS↗

arXiv2024

Weakly Leveled Planarity with Bounded Span

This paper studies planar drawings of graphs in which each vertex is represented as a point along a sequence of horizontal lines, called levels, and each edge is either a horizontal segment or a strictly $y$-monotone curve. A graph is $s$-span weakly leveled planar if it admits such a drawing where the edges have span at most $s$; the span of an edge is the number of levels it touches minus one. We investigate the problem of computing $s$-span weakly leveled planar drawings from both the computational and the combinatorial perspectives. We prove the problem to be para-NP-hard with respect to its natural parameter $s$ and investigate its complexity with respect to widely used structural parameters. We show the existence of a polynomial-size kernel with respect to vertex cover number and prove that the problem is FPT when parameterized by treedepth. We also present upper and lower bounds on the span for various graph classes. Notably, we show that cycle trees, a family of $2$-outerplanar graphs generalizing Halin graphs, are $\Theta(\log n)$-span weakly leveled planar and $4$-span weakly leveled planar when $3$-connected. As a byproduct of these combinatorial results, we obtain improved bounds on the edge-length ratio of the graph families under consideration.

cs.CG↗

arXiv2023

Drawn Tree Decomposition: New Approach for Graph Drawing Problems

Over the past decade, we witness an increasing amount of interest in the design of exact exponential-time and parameterized algorithms for problems in Graph Drawing. Unfortunately, we still lack knowledge of general methods to develop such algorithms. An even more serious issue is that, here, "standard" parameters very often yield intractability. In particular, for the most common structural parameter, namely, treewidth, we frequently observe NP-hardness already when the input graphs are restricted to have constant (often, being just $1$ or $2$) treewidth. Our work deals with both drawbacks simultaneously. We introduce a novel form of tree decomposition that, roughly speaking, does not decompose (only) a graph, but an entire drawing. As such, its bags and separators are of geometric (rather than only combinatorial) nature. While the corresponding parameter -- like treewidth -- can be arbitrarily smaller than the height (and width) of the drawing, we show that -- unlike treewidth -- it gives rise to efficient algorithms. Specifically, we get slice-wise polynomial (XP) time algorithms parameterized by our parameter. We present a general scheme for the design of such algorithms, and apply it to several central problems in Graph Drawing, including the recognition of grid graphs, minimization of crossings and bends, and compaction. Other than for the class of problems we discussed in the paper, we believe that our decomposition and scheme are of independent interest and can be further extended or generalized to suit even a wider class of problems. Additionally, we discuss classes of drawings where our parameter is bounded by $O(\sqrt{n})$ (where $n$ is the number of vertices of the graph), yielding subexponential-time algorithms. Lastly, we prove which relations exist between drawn treewidth and other width measures, including treewidth, pathwidth, (dual) carving-width and embedded-width.

cs.DS↗

arXiv2023

Collective Graph Exploration Parameterized by Vertex Cover

We initiate the study of the parameterized complexity of the {\sc Collective Graph Exploration} ({\sc CGE}) problem. In {\sc CGE}, the input consists of an undirected connected graph $G$ and a collection of $k$ robots, initially placed at the same vertex $r$ of $G$, and each one of them has an energy budget of $B$. The objective is to decide whether $G$ can be \emph{explored} by the $k$ robots in $B$ time steps, i.e., there exist $k$ closed walks in $G$, one corresponding to each robot, such that every edge is covered by at least one walk, every walk starts and ends at the vertex $r$, and the maximum length of any walk is at most $B$. Unfortunately, this problem is \textsf{NP}-hard even on trees [Fraigniaud {\em et~al.}, 2006]. Further, we prove that the problem remains \textsf{W[1]}-hard parameterized by $k$ even for trees of treedepth $3$. Due to the \textsf{para-NP}-hardness of the problem parameterized by treedepth, and motivated by real-world scenarios, we study the parameterized complexity of the problem parameterized by the vertex cover number ($\mathsf{vc}$) of the graph, and prove that the problem is fixed-parameter tractable (\textsf{FPT}) parameterized by $\mathsf{vc}$. Additionally, we study the optimization version of {\sc CGE}, where we want to optimize $B$, and design an approximation algorithm with an additive approximation factor of $O(\mathsf{vc})$.

cs.DS↗

arXiv2023

The Parametrized Complexity of the Segment Number

Given a straight-line drawing of a graph, a segment is a maximal set of edges that form a line segment. Given a planar graph $G$, the segment number of $G$ is the minimum number of segments that can be achieved by any planar straight-line drawing of $G$. The line cover number of $G$ is the minimum number of lines that support all the edges of a planar straight-line drawing of $G$. Computing the segment number or the line cover number of a planar graph is $\exists\mathbb{R}$-complete and, thus, NP-hard. We study the problem of computing the segment number from the perspective of parameterized complexity. We show that this problem is fixed-parameter tractable with respect to each of the following parameters: the vertex cover number, the segment number, and the line cover number. We also consider colored versions of the segment and the line cover number.

cs.CG↗

arXiv2023

On the Exponential Growth of Geometric Shapes

In this paper, we explore how geometric structures can be grown exponentially fast. The studied processes start from an initial shape and apply a sequence of centralized growth operations to grow other shapes. We focus on the case where the initial shape is just a single node. A technical challenge in growing shapes that fast is the need to avoid collisions caused when the shape breaks, stretches, or self-intersects. We identify a parameter $k$, representing the number of turning points within specific parts of a shape. We prove that, if edges can only be formed when generating new nodes and cannot be deleted, trees having $O(k)$ turning points on every root-to-leaf path can be grown in $O(k\log n)$ time steps and spirals with $O(\log n)$ turning points can be grown in $O(\log n)$ time steps, $n$ being the size of the final shape. For this case, we also show that the maximum number of turning points in a root-to-leaf path of a tree is a lower bound on the number of time steps to grow the tree and that there exists a class of paths such that any path in the class with $\Omega(k)$ turning points requires $\Omega(k\log k)$ time steps to be grown. If nodes can additionally be connected as soon as they become adjacent, we prove that if a shape $S$ has a spanning tree with $O(k)$ turning points on every root-to-leaf path, then the adjacency closure of $S$ can be grown in $O(k \log n)$ time steps. In the strongest model that we study, where edges can be deleted and neighbors can be handed over to newly generated nodes, we obtain a universal algorithm: for any shape $S$ it gives a process that grows $S$ from a single node exponentially fast.

cs.DS↗

arXiv2023

Collision Detection for Modular Robots -- it is easy to cause collisions and hard to avoid them

We consider geometric collision-detection problems for modular reconfigurable robots. Assuming the nodes (modules) are connected squares on a grid, we investigate the complexity of deciding whether collisions may occur, or can be avoided, if a set of expansion and contraction operations is executed. We study both discrete- and continuous-time models, and allow operations to be coupled into a single parallel group. Our algorithms to decide if a collision may occur run in $O(n^2\log^2 n)$ time, $O(n^2)$ time, or $O(n\log^2 n)$ time, depending on the presence and type of coupled operations, in a continuous-time model for a modular robot with $n$ nodes. To decide if collisions can be avoided, we show that a very restricted version is already NP-complete in the discrete-time model, while the same problem is polynomial in the continuous-time model. A less restricted version is NP-hard in the continuous-time model.

cs.CG↗

arXiv2022

1st Workshop on Maritime Computer Vision (MaCVi) 2023: Challenge Results

The 1$^{\text{st}}$ Workshop on Maritime Computer Vision (MaCVi) 2023 focused on maritime computer vision for Unmanned Aerial Vehicles (UAV) and Unmanned Surface Vehicle (USV), and organized several subchallenges in this domain: (i) UAV-based Maritime Object Detection, (ii) UAV-based Maritime Object Tracking, (iii) USV-based Maritime Obstacle Segmentation and (iv) USV-based Maritime Obstacle Detection. The subchallenges were based on the SeaDronesSee and MODS benchmarks. This report summarizes the main findings of the individual subchallenges and introduces a new benchmark, called SeaDronesSee Object Detection v2, which extends the previous benchmark by including more classes and footage. We provide statistical and qualitative analyses, and assess trends in the best-performing methodologies of over 130 submissions. The methods are summarized in the appendix. The datasets, evaluation code and the leaderboard are publicly available at https://seadronessee.cs.uni-tuebingen.de/macvi.

cs.CV↗

arXiv2022

Distributed Reconfiguration of Spanning Trees

In a reconfiguration problem, given a problem and two feasible solutions of the problem, the task is to find a sequence of transformations to reach from one solution to the other such that every intermediate state is also a feasible solution to the problem. In this paper, we study the distributed spanning tree reconfiguration problem and we define a new reconfiguration step, called $k$-simultaneous add and delete, in which every node is allowed to add at most $k$ edges and delete at most $k$ edges such that multiple nodes do not add or delete the same edge. We first observe that, if the two input spanning trees are rooted, then we can do the reconfiguration using a single $1$-simultaneous add and delete step in one round in the CONGEST model. Therefore, we focus our attention towards unrooted spanning trees and show that transforming an unrooted spanning tree into another using a single $1$-simultaneous add and delete step requires $Ω(n)$ rounds in the LOCAL model. We additionally show that transforming an unrooted spanning tree into another using a single $2$-simultaneous add and delete step can be done in $O(\log n)$ rounds in the CONGEST model.

cs.DC↗