arXiv subjects

Jinwoo Kim

Publications and source records attributed to Jinwoo Kim.

At least 37 records · Page 2Linked to original sources

arXiv2026

Continuous Diffusion Models Can Obey Formal Syntax

Diffusion language models offer a promising alternative to autoregressive models due to their global, non-causal generation process, but their continuous latent dynamics make discrete constraints -- e.g., the output should be a JSON file that matches a given schema -- difficult to impose. We introduce a training-free guidance method for steering continuous diffusion language models to satisfy formal syntactic constraints expressed using regular expressions. Our approach constructs an analytic score estimating the probability that a latent state decodes to a valid string accepted by a given regular expression, and uses its gradient to guide sampling, without training auxiliary classifiers. The denoising process targets the base model conditioned on syntactic validity. We implement our method in Diffinity on top of the PLAID diffusion model and evaluate it on 180 regular-expression constraints over JSON and natural-language benchmarks. Diffinity achieves 68-96\% constraint satisfaction while incurring only a small perplexity cost relative to unconstrained sampling, outperforming autoregressive constrained decoding in both constraint satisfaction and output quality. Diffinity is open-sourced at github.com/large-loris-models/Diffinity.

cs.LG↗

arXiv2026

Learning the Error Patterns of Language Models

When generating outputs for domains with specific validity constraints (e.g., a program should compile), LLMs often fail in a small number of focused ways: for example, by using Python function names when generating TypeScript. We observe that these error patterns can be represented using a small number of constraints that can be learned in practice. We propose \emph{prefix filters}, which are per-domain-and-LLM symbolic functions, as objects to capture the error patterns, Palla as an algorithm to learn prefix filters efficiently in practice, and implement Palla. Prefix filters learned by Palla i) help us quantitatively analyze the error patterns of LLMs, and ii) can be used to constrain the outputs of a model via constrained sampling algorithms. For example, Palla boosts compile rates for Qwen2.5-1.5B on TypeScript generation, by over 60%, allowing Qwen2.5-1.5B to achieve similar performance to Llama3.1-8B unconstrained.

cs.LG↗

arXiv2026

What are the Right Symmetries for Formal Theorem Proving?

Formal theorem provers based on large language models (LLMs) are highly sensitive to superficial variations in problem representation: semantically equivalent statements can exhibit drastically different proof success rates, revealing a failure to respect structural symmetries inherent in formal mathematics. This raises a central question: what are the right symmetries for formal theorem proving? We introduce rewriting categories, a category-theoretic framework capturing the compositional, generally non-invertible transformations induced by proof tactics, and use it to formalize two symmetry notions: proof equivariance, governing how proof distributions transform under rewrites, and success invariance (i.e., invariance of success probability), requiring equivalent statements to be solved with the same probability. We observe that state-based next-tactic provers naturally satisfy proof equivariance by operating on proof states. In contrast, state-of-the-art LLM-based provers satisfy neither property, exhibiting large performance variation across equivalent formulations. To mitigate this, we propose test-time methods that aggregate over equivalent rewritings of the input, showing theoretically that they recover success invariance in the sampling limit, and empirically, that they improve robustness and performance under fixed inference budgets. Our results highlight symmetry as a key missing inductive bias in LLM-based theorem proving and suggest test-time computation as a practical route to approximate it.

cs.LG↗

arXiv2026

Flow Map Language Models: One-step Language Modeling via Continuous Denoising

Language models based on discrete diffusion have attracted widespread interest for their potential to provide faster generation than autoregressive models. Despite their promise, these models typically produce samples whose quality sharply degrades in the few-step regime, preventing a dramatic speedup in practice. Here, we show that language models based on continuous flows over one-hot token embeddings can outperform discrete diffusion in both quality and speed. Importantly, our continuous formulation defines a unique flow map that can be learned directly for efficient few-step inference, a structure we show is unavailable to discrete methods. In this setting, we show that both the flow and its associated flow map can be learned with simple cross-entropy objectives that respect the simplex geometry of the data, and we identify three distinct choices for flow map distillation whose performance we compare in practice. Using these insights, we build a flow language model (FLM), a continuous flow that matches state-of-the-art discrete diffusion baselines on the One Billion Words (LM1B) and OpenWebText (OWT) datasets. We then distill FLM into a flow map language model (FMLM), whose one-step generation exceeds the 8-step quality of recent few-step discrete diffusion language models. Our work challenges the widely-held hypothesis that discrete noising processes are necessary for generative modeling over discrete modalities and paves the way toward accelerated language modeling at scale. Code is available at https://github.com/david3684/flm.

cs.CL↗

arXiv2026

MUX: Continuous Reasoning via Multiplexed Tokens

Language models solve complex problems by articulating intermediate reasoning steps in natural language. While effective, this process is computationally bottlenecked: each reasoning step conveys only a single subword, and many are spent expressing a thought instead of carrying out computation. We propose MUX, a simple method for high-bandwidth and compact reasoning based on distillation of discrete reasoning into continuous multiplexed tokens in a latent space. Here, each latent token is trained to represent a weighted linear superposition (multiplexing) of a span of discrete reasoning subwords, where this superposition is lossless by construction and the span can be fully recovered (demultiplexing). We prove that simple position-dependent weightings, such as suitable geometric decay, support lossless multiplexing, which in turn prevents shortcut behaviors caused by latent collapse. We further show that multiplexed reasoning can perform parallel exploration in problems that require search. Across 32 evaluation settings spanning four language models, MUX outperforms strong latent reasoning baselines. Ablation and probing analyses further show that the learned latent tokens encode faithful and interpretable reasoning. Our results suggest that lossless superposition as local learning targets constitutes a sufficient condition for achieving strong and efficient latent continuous reasoning.

cs.AI↗

arXiv2026

Data-Driven Monitoring and Deterrence in a Changing Environment

We study a dynamic model in which a principal monitors agents based on historical data of infractions. This data informs when and at what intensity to monitor; the monitoring decision, in turn, selects the collected data, shaping the principal's future learning. We analyze this feedback loop using a bandit model in which the underlying monitoring environment evolves according to a hidden Markov process. Because data collection is endogenous, how the principal uses this information is critical: surprisingly, a myopic approach renders historical data completely valueless. By endogenizing the agent's incentives, we demonstrate that the principal's purely informational motive to explore serves as an endogenous commitment device. This inherent drive to gather data compels persistent vigilance, strictly lowering the equilibrium infraction rate and restoring the power of deterrence.

econ.TH↗

arXiv2026

Flock: A Knowledge Graph Foundation Model via Learning on Random Walks

We study the problem of zero-shot link prediction on knowledge graphs (KGs), which requires models to generalize to novel entities and novel relations. Knowledge graph foundation models (KGFMs) address this task by enforcing equivariance over both nodes and relations, which enables them to learn structural properties of nodes and relations that transfer to novel KGs with similar structure. However, the conventional notion of deterministic equivariance inherently limits the expressive power of KGFMs, as it prevents them from distinguishing relations that are structurally similar but semantically distinct. To overcome this limitation, we propose to leverage probabilistic node-relation equivariance, which preserves equivariance in distribution while using structured randomness to break symmetries at inference time. Building on this principle, we present Flock, a KGFM that iteratively samples random walks, encodes them into sequences, embeds them with a sequence model, and aggregates node and relation representations through learned pooling. Flock respects probabilistic node-relation equivariance and, crucially, is a universal approximator for isomorphism-invariant link-level functions over KGs. Empirically, Flock perfectly solves our new diagnostic dataset Petals on which current KGFMs fail, and achieves state-of-the-art performance on entity and relation prediction tasks across 54 KGs from diverse domains. Code is available at https://github.com/jw9730/flock.

cs.LG↗

arXiv2026

RelAgent: LLM Agents as Data Scientists for Relational Learning

Relational learning is a challenging problem that has motivated a wide range of approaches, including graph-based models (e.g., graph neural networks, graph transformers), tabular methods (e.g., tabular foundation models), and sequence-based approaches (e.g., large language models), each with its own advantages and limitations. We propose RelAgent, an LLM-based autonomous data scientist for relational learning, which operates in two phases. In the search phase, an LLM agent uses database, validation, and evaluation workspace tools to construct SQL feature programs and select a predictive model. In the inference phase, the resulting program is executed without further LLM calls. The final predictor consists of SQL queries and a classical model, enabling fast, deterministic, and intrinsically interpretable predictions: features are human-readable queries, and predictions depend only on the resulting query-defined feature map, enabling scalable deployment using standard database systems.

cs.LG↗

arXiv2026

The Fourth Challenge on Image Super-Resolution ($\times$4) at NTIRE 2026: Benchmark Results and Method Overview

This paper presents the NTIRE 2026 image super-resolution ($\times$4) challenge, one of the associated competitions of the NTIRE 2026 Workshop at CVPR 2026. The challenge aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs generated through bicubic downsampling with a $\times$4 scaling factor. The objective is to develop effective super-resolution solutions and analyze recent advances in the field. To reflect the evolving objectives of image super-resolution, the challenge includes two tracks: (1) a restoration track, which emphasizes pixel-wise fidelity and ranks submissions based on PSNR; and (2) a perceptual track, which focuses on visual realism and evaluates results using a perceptual score. A total of 194 participants registered for the challenge, with 31 teams submitting valid entries. This report summarizes the challenge design, datasets, evaluation protocol, main results, and methods of participating teams. The challenge provides a unified benchmark and offers insights into current progress and future directions in image super-resolution.

cs.CV↗

arXiv2026

HarassGuard: Detecting Harassment Behaviors in Social Virtual Reality with Vision-Language Models

Social Virtual Reality (VR) platforms provide immersive social experiences but also expose users to serious risks of online harassment. Existing safety measures are largely reactive, while proactive solutions that detect harassment behavior during an incident often depend on sensitive biometric data, raising privacy concerns. In this paper, we present HarassGuard, a vision-language model (VLM) based system that detects physical harassment in social VR using only visual input. We construct an IRB-approved harassment vision dataset, apply prompt engineering, and fine-tune VLMs to detect harassment behavior by considering contextual information in social VR. Experimental results demonstrate that HarassGuard achieves competitive performance compared to state-of-the-art baselines (i.e., LSTM/CNN, Transformer), reaching an accuracy of up to 88.09% in binary classification and 68.85% in multi-class classification. Notably, HarassGuard matches these baselines while using significantly fewer fine-tuning samples (200 vs. 1,115), offering unique advantages in contextual reasoning and privacy-preserving detection.

cs.CV↗

arXiv2026

Near-Real-Time InSAR Phase Estimation for Large-Scale Surface Displacement Monitoring

Operational near-real-time monitoring of Earth's surface deformation using Interferometric Synthetic Aperture Radar (InSAR) requires processing algorithms that efficiently incorporate new acquisitions without reprocessing historical archives. We present sequential phase linking approach using compressed single-look-complex images (SLCs) capable of producing surface displacement estimates within hours of the time of a new acquisition. Our key algorithmic contribution is a mini-stack reference scheme that maintains phase consistency across processing batches without adjusting or re-estimating previous time steps, enabling straightforward operational deployment. We introduce online methods for persistent and distributed scatterer identification that adapt to temporal changes in surface properties through incremental amplitude statistics updates. The processing chain incorporates multiple complementary metrics for pixel quality that are reliable for small SLC stack sizes, and an L1-norm network inversion to limit propagation of unwrapping errors across the time series. We use our algorithm to produce OPERA Surface Displacement from Sentinel-1 product, the first continental-scale surface displacement product over North America. Validation against GPS measurements and InSAR residual analysis demonstrates millimeter-level agreement in velocity estimates in varying environmental conditions. We demonstrate our algorithm's capabilities with a successful recovery of meter-scale co-eruptive displacement at Kilauea volcano during the 2018 eruption, as well as detection of subtle uplift at Three Sisters volcano, Oregon -- a challenging environment for C-band InSAR due to dense vegetation and seasonal snow. We have made all software available as open source libraries, providing a significant advancement to the open scientific community's ability to process large InSAR data sets in a cloud environment.

eess.SP↗

arXiv2026

Monotone Comparative Statics without Lattices

The theory of Monotone Comparative Statics (MCS) has traditionally required a lattice structure, excluding certain multidimensional environments such as mixed-strategy games where this property fails. We show that this structure is not essential. We introduce a weaker notion, the pseudo-lattice property, and preserve the theory's core results by generalizing the MCS theorems for individual choice and Tarski's fixed-point theorem. Our framework expands comparative statics to pseudo quasi-supermodular games. Crucially, it enables the first MCS analysis of mixed-strategy Nash equilibria and trembling-hand perfect equilibria.

econ.TH↗

arXiv2025

Investigating the Integrated Digital Interventions Delivered by a Therapeutic Companion Agent for Young Adults with Symptoms of Depression: A Proof-of-Concept Study

Background: Despite the clinical effectiveness of digital interventions for young adults with depression, low engagement and adherence remain persistent challenges. Building a strong digital therapeutic alliance has been proposed to address these barriers. This study highlights the need for a conversational therapeutic companion agent (TCA)-based intervention design. Objective: This study aimed to develop a Wizard-of-Oz TCA-centered prototype integrating social-support-based ecological momentary assessment (EMA), ecological momentary intervention (EMI), behavioral activation, and gamification. We evaluated the six-week proof-of-concept efficacy of this intervention among young adults with depressive symptoms. Methods: Korean young adults aged 20--39 years with mild-to-moderate depressive symptoms (PHQ-9) were recruited online. The intervention group ($n = 29$) received a six-week TCA-based digital intervention, while the control group ($n = 29$), recruited four weeks later, continued their usual routines. The TCA guided four daily behavioral-activation tasks, three mood assessments, meditation, daily summaries, and weekly mission feedback. Both groups were assessed at baseline and at weeks 2, 4, and 6 using the BDI-II, GAD-7, and Q-LES-Q-SF. Results: Of 58 participants, 57 completed the study (one dropout in the intervention group). At week 6, the intervention group showed significantly greater reductions in depressive symptoms and improvements in quality of life than controls. Adherence was 78\% for EMA, 51\% for EMI, and 65\% for daily routines. Conclusions: The TCA-based digital intervention improved depressive symptoms and quality of life with adherence levels comparable to previous digital health interventions. Future studies should refine the TCA design and conduct larger-scale evaluations.

cs.HC↗

arXiv2025

BEACON: Automatic Container Policy Generation using Environment-aware Dynamic Analysis

This paper introduces BeaCon, a novel tool for the automated generation of adjustable container security policies. Unlike prior approaches, BeaCon leverages dynamic analysis to simulate realistic environments, uncovering container execution paths that may remain hidden during the profiling phase. To address the challenge of exploring vast profiling spaces, we employ efficient heuristics to reveal additional system events with minimal effort. In addition, BeaCon incorporates a security and functionality scoring mechanism to prioritize system calls and capabilities based on their impact on the host OS kernel's security and the functionality of containerized applications. By integrating these scores, BeaCon achieves a customized balance between security and functionality, enabling cloud providers to enforce security measures while maintaining tenant availability. We implemented a prototype of BeaCon using eBPF kernel technology and conducted extensive evaluations. Results from the top 15 containers, which revealed significant improvements, demonstrate that BeaCon identifies an average of 16.5% additional syscalls by applying diverse environments. Furthermore, we evaluated its effectiveness in mitigating risks associated with 45 known vulnerabilities (e.g., CVEs), showcasing its potential to significantly enhance container security. Additionally, we performed proof-of-concept demonstrations for two well-known security vulnerabilities, showing that BeaCon successfully reduces attack surface by blocking these exploits.

cs.CR↗

arXiv2025

SecTracer: A Framework for Uncovering the Root Causes of Network Intrusions via Security Provenance

Modern enterprise networks comprise diverse and heterogeneous systems that support a wide range of services, making it challenging for administrators to track and analyze sophisticated attacks such as advanced persistent threats (APTs), which often exploit multiple vectors. To address this challenge, we introduce the concept of network-level security provenance, which enables the systematic establishment of causal relationships across hosts at the network level, facilitating the accurate identification of the root causes of security incidents. Building on this concept, we present SecTracer as a framework for a network-wide provenance analysis. SecTracer offers three main contributions: (i) comprehensive and efficient forensic data collection in enterprise networks via software-defined networking (SDN), (ii) reconstruction of attack histories through provenance graphs to provide a clear and interpretable view of intrusions, and (iii) proactive attack prediction using probabilistic models. We evaluated the effectiveness and efficiency of SecTracer through a real-world APT simulation, demonstrating its capability to enhance threat mitigation while introducing less than 1% network throughput overhead and negligible latency impact.

cs.CR↗

arXiv2025

CryptoGuard: Lightweight Hybrid Detection and Response to Host-based Cryptojackers in Linux Cloud Environments

Host-based cryptomining malware, commonly known as cryptojackers, have gained notoriety for their stealth and the significant financial losses they cause in Linux-based cloud environments. Existing solutions often struggle with scalability due to high monitoring overhead, low detection accuracy against obfuscated behavior, and lack of integrated remediation. We present CryptoGuard, a lightweight hybrid solution that combines detection and remediation strategies to counter cryptojackers. To ensure scalability, CryptoGuard uses sketch- and sliding window-based syscall monitoring to collect behavior patterns with minimal overhead. It decomposes the classification task into a two-phase process, leveraging deep learning models to identify suspicious activity with high precision. To counter evasion techniques such as entry point poisoning and PID manipulation, CryptoGuard integrates targeted remediation mechanisms based on eBPF, a modern Linux kernel feature deployable on any compatible host. Evaluated on 123 real-world cryptojacker samples, it achieves average F1-scores of 96.12% and 92.26% across the two phases, and outperforms state-of-the-art baselines in terms of true and false positive rates, while incurring only 0.06% CPU overhead per host.

cs.CR↗

arXiv2025

The Hidden Dangers of Public Serverless Repositories: An Empirical Security Assessment

Serverless computing has rapidly emerged as a prominent cloud paradigm, enabling developers to focus solely on application logic without the burden of managing servers or underlying infrastructure. Public serverless repositories have become key to accelerating the development of serverless applications. However, their growing popularity makes them attractive targets for adversaries. Despite this, the security posture of these repositories remains largely unexplored, exposing developers and organizations to potential risks. In this paper, we present the first comprehensive analysis of the security landscape of serverless components hosted in public repositories. We analyse 2,758 serverless components from five widely used public repositories popular among developers and enterprises, and 125,936 Infrastructure as Code (IaC) templates across three widely used IaC frameworks. Our analysis reveals systemic vulnerabilities including outdated software packages, misuse of sensitive parameters, exploitable deployment configurations, susceptibility to typo-squatting attacks and opportunities to embed malicious behaviour within compressed serverless components. Finally, we provide practical recommendations to mitigate these threats.

cs.CR↗

arXiv2025

Heimdallr: Fingerprinting SD-WAN Control-Plane Architecture via Encrypted Control Traffic

Software-defined wide area network (SD-WAN) has emerged as a new paradigm for steering a large-scale network flexibly by adopting distributed software-defined network (SDN) controllers. The key to building a logically centralized but physically distributed control-plane is running diverse cluster management protocols to achieve consistency through an exchange of control traffic. Meanwhile, we observe that the control traffic exposes unique time-series patterns and directional relationships due to the operational structure even though the traffic is encrypted, and this pattern can disclose confidential information such as control-plane topology and protocol dependencies, which can be exploited for severe attacks. With this insight, we propose a new SD-WAN fingerprinting system, called Heimdallr. It analyzes periodical and operational patterns of SD-WAN cluster management protocols and the context of flow directions from the collected control traffic utilizing a deep learning-based approach, so that it can classify the cluster management protocols automatically from miscellaneous control traffic datasets. Our evaluation, which is performed in a realistic SD-WAN environment consisting of geographically distant three campus networks and one enterprise network shows that Heimdallr can classify SD-WAN control traffic with $\geq$ 93%, identify individual protocols with $\geq$ 80% macro F-1 scores, and finally can infer control-plane topology with $\geq$ 70% similarity.

cs.CR↗