arXiv subjects

Sergey Ivanov

Publications and source records attributed to Sergey Ivanov.

At least 19 recordsLinked to original sources

arXiv2026

The Class Edge-Reconstruction Number of a Maximal Planar Graph Is One or Two

An edge card of a graph is obtained by deleting one edge, and a class edge-reconstruction number asks for the fewest carefully selected cards that identify the graph when its class is known. We determine the sharp universal bound for maximal planar graphs. Two selected cards always suffice, and the octahedral graph shows that two can be necessary; some maximal planar graphs are already identified by one card. The argument exploits the fact that deleting a flippable edge leaves a single quadrilateral whose two diagonals give the only possible maximal-planar completions. Degree information then rules out the competing completion, with a separate argument for graphs containing a vertex of degree three. This settles a problem posed in a 2010 survey on reconstruction numbers.

math.CO↗

arXiv2026

Graphs with Long Pseudosimilarity Chains under Consecutive Vertex Deletions

Pseudosimilar vertices are vertices in distinct automorphism orbits whose deletions produce isomorphic graphs. Classical work has studied the existence, group-theoretic origin, and construction of large sets of such vertices. We ask a different recursive question: how long can one repeatedly delete a vertex that is pseudosimilar at the moment of deletion? We define the pseudosimilarity depth of a graph and construct connected graphs in which this process continues through all but a sublinear number of vertices. A two-clock construction gives a square-root deficit uniformly in the order, while a Chinese-remainder construction with many cyclic clocks yields an infinite family of asymmetric graphs with only a polylogarithmic number of vertices left outside the active chain. The mechanism realizes pseudosimilarity by breaking a long hidden automorphism orbit and enlarging the break one vertex at a time. Thus pseudosimilarity can persist through an asymptotically full sequence of vertex deletions, even though every graph encountered in the main construction is asymmetric.

math.CO↗

arXiv2026

Cost-Effective Repository Exploration for Agentic Issue Localization

Repository exploration is a distinct and costly stage of coding-agent pipelines: before generating a patch, an agent must identify which repository files are likely to matter. We study whether this stage can be delegated to lower-cost models while retaining useful localization quality. Using our IssueLoc-Bench, we evaluate five explorer models under the same read-only interactive interface on 499 SWE-bench Verified-derived tasks and 500 tasks from 153 additional repositories. We measure early candidate discovery, top-three gold-file coverage, strict file-set recovery, agent time, and token usage, with paired instance-level uncertainty and repository-clustered sensitivity analysis. The highest-quality explorer leads across localization metrics, but substantially cheaper operating points emerge: depending on the model and evaluation arm, lower-cost explorers retain approximately 78-94% of the reference Hit@3 and 73-92% of its F1 while reducing mean agent time by 41-88% and token usage by 84-95%. The preferred operating point depends on how localization is consumed downstream: ranking and coverage metrics characterize recoverable candidate handoffs, whereas F1 and exact match characterize restrictive file gates. These results support treating repository exploration as an independently measurable and budgetable stage of modular coding agents, with explorer selection guided by the downstream handoff contract.

cs.SE↗

arXiv2026

Spec2Vision: Contract-Guided Delivery of AI-Generated Computer Vision Pipelines

Generated computer-vision code can be runnable without satisfying the task contract enforced by a downstream evaluator. We study that gap with Spec2Vision, an experimental framework for producing and evaluating specification-grounded CV pipeline bundles through a staged runtime that keeps the task contract explicit across synthesis, screening, testing, and bounded repair. The benchmark evaluates 17 CV tasks, 10 executable conditions, and 5 repeats per task-condition cell, for 850 primary runs. In the primary 850-run evaluation, Spec2Vision reaches 81/85 evaluator-test passes; removing structural repair drops to 55/85, compatibility scaffolding to 58/85, and generator preflight to 39/85. The executable single-agent baselines expose progressively richer task specifications to the model, culminating in direct source-spec exposure, yet remain much weaker overall, from 17/85 for lightweight task grounding to 35/85 evaluator-test passes. The lightweight baseline nevertheless remains core-runnable in 85/85 runs but reaches only 17/85 evaluator-test passes and 6/85 strict-delivery successes, showing that runnability is not equivalent to delivery. Across this benchmark, the strongest evidence comes from keeping the task contract explicit across staged generation, checking, and repair. Artifacts are provided to support audit of run bundles, model-visible inputs, and derived tables.

cs.SE↗

arXiv2026

Nonisomorphic Graphs Can Share an Arbitrarily Large Fraction of Their Vertex-Deleted Cards

For a graph $G$, its vertex deck is the multiset of graphs obtained by deleting one vertex. Bowler, Brown, and Fenner (BBF) proposed $2\lfloor(n-1)/3\rfloor$ as the maximum possible overlap between the decks of two nonisomorphic $n$-vertex graphs, for all sufficiently large $n$. We first give an explicit pair of connected nonisomorphic graphs on $78$ vertices with at least $51$ common cards, exceeding BBF's predicted value of $50$. We then construct, for every even $r\ge4$, families at arbitrarily large orders whose overlap fraction is asymptotically at least $1-1/r$. Consequently, for every $\alpha<1$, infinitely many pairs have more than $\alpha n$ common cards, so the attainable fraction is arbitrarily close to the full deck. For representative instances, the predicted overlaps were also checked by complete deck generation and isomorphism testing with Brendan McKay's nauty tools.

math.CO↗

arXiv2026

IDP AutoOpt: Agent-Driven Optimization of Document Processing Pipeline Configurations

We present IDP AutoOpt, an autonomous LLM agent that discovers high-performing configurations for intelligent document processing (IDP) pipelines. Tuning IDP prompts, models, OCR settings, and schemas jointly currently costs domain specialists 20 to 80+ person-hours per document type and does not scale as enterprises add document classes. IDP AutoOpt runs a closed loop: it scores a configuration on a small labeled set, diagnoses field-level errors, generates targeted edits, and re-evaluates, guided by human-authored domain skills that encode production expertise. Across extraction, classification, and packet-splitting tasks deployed in healthcare, marketing-intelligence, and financial-services settings, IDP AutoOpt matches or exceeds human-expert accuracy at equal or lower cost (on an extraction benchmark, 90.2% vs 81.6% at 4.6 x lower per-page cost), cutting configuration time from weeks to under two hours. We further show that agent LLM capability has a hard threshold below which optimization fails, and that curated domain skills outperform raw source-code access, which can degrade performance when provided without structure. We also share practical lessons on context management and variance mitigation. Requiring only a configurable pipeline, a scoring function, and a small labeled set, the approach extends beyond IDP to other enterprise AI systems, such as RAG and multi-agent workflows, where configuration bottlenecks deployment.

cs.IR↗

arXiv2026

Loc2Repair: A Framework for Evaluating the Impact of File-Level Issue Localization in Repo-Level LLM Repair

Repository-grounded automated repair is often reported as a single end-to-end capability, which hides distinct failure modes such as poor file targeting, incorrect patch synthesis, and failed iterative debugging. We present Loc2Repair, a modular evaluation framework for controlled analysis of repository-grounded repair pipelines, and use it to isolate file-level issue localization as an upstream variable. Loc2Repair decouples localization and repair under a shared runtime, artifact schema, and evaluation harness, allowing researchers to combine different localization models and repair backbones under matched conditions. Using three repair backbones on SWE-bench Verified, we compare baseline repair without explicit localization, repair guided by predicted localization from two localizers, and repair guided by gold modified-file sets. Explicit localization consistently improves resolved rate across all backbones: pooled performance increases from 44.7% for baseline repair to 48.9% and 49.1% with predicted localization, and to 52.4% with gold localization. Localization also reduces mean elapsed time overall: in pooled paired analysis, mean elapsed time decreases by 100.94 s and 52.25 s for the two predicted-localization settings, and by 154.45 s with gold guidance, although token effects remain heterogeneous across models. Overall, Loc2Repair shows file-level localization is a consistent repair lever, improving effectiveness and mean latency in pooled analysis, while gold-guided failures expose headroom beyond localization.

cs.SE↗

arXiv2025

Pre-Filtering Code Suggestions using Developer Behavioral Telemetry to Optimize LLM-Assisted Programming

Large Language Models (LLMs) are increasingly integrated into code editors to provide AI-powered code suggestions. Yet many of these suggestions are ignored, resulting in wasted computation, increased latency, and unnecessary interruptions. We introduce a lightweight pre-filtering model that predicts the likelihood of suggestion acceptance before invoking the LLM, using only real-time developer telemetry such as typing speed, file navigation, and editing activity. Deployed in a production-grade Visual Studio Code plugin over four months of naturalistic use, our approach nearly doubled acceptance rates (18.4% -> 34.2%) while suppressing 35% of low-value LLM calls. These findings demonstrate that behavioral signals alone can meaningfully improve both user experience and system efficiency in LLM-assisted programming, highlighting the value of timing-aware, privacy-preserving adaptation mechanisms. The filter operates solely on pre-invocation editor telemetry and never inspects code or prompts.

cs.SE↗

arXiv2025

Optimizing LLM Code Suggestions: Feedback-Driven Timing with Lightweight State Bounds

Large Language Models (LLMs) have transformed code auto-completion by generating context-aware suggestions. Yet, deciding when to present these suggestions remains underexplored, often leading to interruptions or wasted inference calls. We propose an adaptive timing mechanism that dynamically adjusts the delay before offering a suggestion based on real-time developer feedback. Our suggested method combines a logistic transform of recent acceptance rates with a bounded delay range, anchored by a high-level binary prediction of the developer's cognitive state. In a two-month deployment with professional developers, our system improved suggestion acceptance from 4.9% with no delay to 15.4% with static delays, and to 18.6% with adaptive timing-while reducing blind rejections (rejections without being read) from 8.3% to 0.36%. Together, these improvements increase acceptance and substantially reduce wasted inference calls by 75%, making LLM-based code assistants more efficient and cost-effective in practice.

cs.SE↗

arXiv2025

A Multimodal Conversational Agent for Tabular Data Analysis

Large language models (LLMs) can reshape information processing by handling data analysis, visualization, and interpretation in an interactive, context-aware dialogue with users, including voice interaction, while maintaining high performance. In this article, we present Talk2Data, a multimodal LLM-driven conversational agent for intuitive data exploration. The system lets users query datasets with voice or text instructions and receive answers as plots, tables, statistics, or spoken explanations. Built on LLMs, the suggested design combines OpenAI Whisper automatic speech recognition (ASR) system, Qwen-coder code generation LLM/model, custom sandboxed execution tools, and Coqui library for text-to-speech (TTS) within an agentic orchestration loop. Unlike text-only analysis tools, it adapts responses across modalities and supports multi-turn dialogues grounded in dataset context. In an evaluation of 48 tasks on three datasets, our prototype achieved 95.8% accuracy with model-only generation time under 1.7 seconds (excluding ASR and execution time). A comparison across five LLM sizes (1.5B-32B) revealed accuracy-latency-cost trade-offs, with a 7B model providing the best balance for interactive use. By routing between conversation with user and code execution, constrained to a transparent sandbox, with simultaneously grounding prompts in schema-level context, the Talk2Data agent reliably retrieves actionable insights from tables while making computations verifiable. In the article, except for the Talk2Data agent itself, we discuss implications for human-data interaction, trust in LLM-driven analytics, and future extensions toward large-scale multimodal assistants.

cs.AI↗

arXiv2022

Simplifying Node Classification on Heterophilous Graphs with Compatible Label Propagation

Graph Neural Networks (GNNs) have been predominant for graph learning tasks; however, recent studies showed that a well-known graph algorithm, Label Propagation (LP), combined with a shallow neural network can achieve comparable performance to GNNs in semi-supervised node classification on graphs with high homophily. In this paper, we show that this approach falls short on graphs with low homophily, where nodes often connect to the nodes of the opposite classes. To overcome this, we carefully design a combination of a base predictor with LP algorithm that enjoys a closed-form solution as well as convergence guarantees. Our algorithm first learns the class compatibility matrix and then aggregates label predictions using LP algorithm weighted by class compatibilities. On a wide variety of benchmarks, we show that our approach achieves the leading performance on graphs with various levels of homophily. Meanwhile, it has orders of magnitude fewer parameters and requires less execution time. Empirical evaluations demonstrate that simple adaptations of LP can be competitive in semi-supervised node classification in both homophily and heterophily regimes.

cs.LG↗

arXiv2022

Reinforcement Learning Textbook

This textbook covers principles behind main modern deep reinforcement learning algorithms that achieved breakthrough results in many domains from game AI to robotics. All required theory is explained with proofs using unified notation and emphasize on the differences between different types of algorithms and the reasons why they are constructed the way they are.

cs.LG↗

arXiv2021

Combining Reward and Rank Signals for Slate Recommendation

We consider the problem of slate recommendation, where the recommender system presents a user with a collection or slate composed of K recommended items at once. If the user finds the recommended items appealing then the user may click and the recommender system receives some feedback. Two pieces of information are available to the recommender system: was the slate clicked? (the reward), and if the slate was clicked, which item was clicked? (rank). In this paper, we formulate several Bayesian models that incorporate the reward signal (Reward model), the rank signal (Rank model), or both (Full model), for non-personalized slate recommendation. In our experiments, we analyze performance gains of the Full model and show that it achieves significantly lower error as the number of products in the catalog grows or as the slate size increases.

cs.LG↗

arXiv2021

Solar activity classification based on Mg II spectra: towards classification on compressed data

Although large volumes of solar data are available for study, the vast majority of these data remain unlabeled and are therefore not amenable to supervised machine learning methods. Having a way to accurately and automatically classify spectra into categories related to solar activity is highly desirable and will assist and speed up future research efforts in solar physics. At the same time, the large volume of raw observational data is a serious bottleneck for machine learning, requiring powerful computational means that are not at the disposal of many laboratories. Besides, the raw data communication imposes restrictions on real time data observations and requires considerable bandwidth and energy for the onboard solar observation systems. To solve these issues, we propose a framework to classify solar activity on compressed data. For this, we used a labeling scheme from a pre-existing vector quantization technique in conjunction with different machine learning algorithms to categorize spectra of singly-ionized magnesium Mg II measured by NASA's Interface Region Imaging Spectrograph satellite (IRIS) into five types of solar activity. Our training dataset is a human annotated list of 85 IRIS observations containing 29097 frames. The annotated types of Solar activities are active region, pre-flare activity, Solar flare, Sunspot, and quiet Sun. We compress these data and reduce its complexity before training classifiers. We found that the XGBoost classifier produces the most accurate results on the compressed data, yielding over a 95\% prediction rate, and outperforming other ML methods like convolution neural networks, K-nearest neighbors, naive Bayes classifiers, and SVM. We find that the classification performance on compressed and uncompressed data is comparable, implying the possibility of large compression rates for relatively low degrees of information loss.

astro-ph.SR↗

arXiv2021

A nonlinear metasurface combining telecom-range intersubband transitions in GaN/AlN quantum wells with resonant plasmonic antenna arrays

We realize and investigate a nonlinear metasurface taking advantage of intersubband transitions in ultranarrow GaN/AlN multi-quantum well heterostructures. Owing to huge band offsets, the structures offer resonant transitions in the telecom window around 1.55 $μ$m. These heterostructures are functionalized with an array of plasmonic antennas featuring cross-polarized resonances at these near-infrared wavelengths and their second harmonic. This kind of nonlinear metasurface allows for substantial second-harmonic generation at normal incidence which is completely absent for an antenna array without the multi-quantum well structure underneath. While the second harmonic is originally radiated only into the plane of the quantum wells, a proper geometrical arrangement of the plasmonic elements permits to redirect the second-harmonic light to free-space radiation, which is emitted perpendicular to the surface.

physics.optics↗

arXiv2020

Are Hyperbolic Representations in Graphs Created Equal?

Recently there was an increasing interest in applications of graph neural networks in non-Euclidean geometry; however, are non-Euclidean representations always useful for graph learning tasks? For different problems such as node classification and link prediction we compute hyperbolic embeddings and conclude that for tasks that require global prediction consistency it might be useful to use non-Euclidean embeddings, while for other tasks Euclidean models are superior. To do so we first fix an issue of the existing models associated with the optimization process at zero curvature. Current hyperbolic models deal with gradients at the origin in ad-hoc manner, which is inefficient and can lead to numerical instabilities. We solve the instabilities of kappa-Stereographic model at zero curvature cases and evaluate the approach of embedding graphs into the manifold in several graph representation learning tasks.

cs.LG↗

arXiv2019

The PanEDM Neutron Electric Dipole Moment Experiment at the ILL

The neutron's permanent electric dipole moment $d_n$ is constrained to below $3\times10^{-26} e~\text{cm}$ (90% C.L.) [ arXiv:hep-ex/0602020, arXiv:1509.04411 ], by experiments using ultracold neutrons (UCN). We plan to improve this limit by an order of magnitude or more with PanEDM, the first experiment exploiting the ILL's new UCN source SuperSUN. SuperSUN is expected to provide a high density of UCN with energies below 80 neV, implying extended statistical reach with respect to existing sources, for experiments that rely on long storage or spin-precession times. Systematic errors in PanEDM are strongly suppressed by passive magnetic shielding, with magnetic field and gradient drifts at the single fT level. A holding-field homogeneity on the order of $10^{-4}$ is achieved in low residual fields, via a high static damping factor and built-in coil system. No comagnetometer is needed for the first order-of-magnitude improvement in $d_n$, thanks to high magnetic stability and an assortment of sensors outside the UCN storage volumes. PanEDM will be commissioned and upgraded in parallel with SuperSUN, to take full advantage of the source's output in each phase. Commissioning is ongoing in 2019, and a new limit in the mid $10^{-27} e~\text{cm}$ range should be possible with two full reactor cycles of data in the commissioned apparatus.

physics.ins-det↗

arXiv2019

Modern Deep Reinforcement Learning Algorithms

Recent advances in Reinforcement Learning, grounded on combining classical theoretical results with Deep Learning paradigm, led to breakthroughs in many artificial intelligence tasks and gave birth to Deep Reinforcement Learning (DRL) as a field of research. In this work latest DRL algorithms are reviewed with a focus on their theoretical justification, practical limitations and observed empirical properties.

cs.LG↗