arXiv subjects

Sulaiman Vesal

Publications and source records attributed to Sulaiman Vesal.

At least 19 recordsLinked to original sources

arXiv2026

AIM: A Privacy-Aware Interoperable Memory Framework for Multi-Agent Multi-User LLM Systems

Traditional large language models (LLMs) are scoped to individual user sessions, limiting their knowledge to a single conversation and preventing them from learning user preferences that evolve over time. Existing agentic memory systems address this limitation but generally operate at the individual-user level, restricting the public knowledge that could be shared across users to improve downstream responses. We introduce AIM (Agentic Interoperable Memory), a unified, privacy-aware memory framework that enables multi-agent, multi-user LLM systems to persistently manage private and shared memory. AIM dynamically classifies information as private, scoped to one user and inaccessible to others, or public, accessible to all users. It enforces index-level access controls so that private memories are retrievable only by their owner, protecting sensitive data while allowing beneficial shared knowledge to improve coordination and consistency. We also introduce MUMBench (Multi-User Memory Benchmark), a dataset of multi-user interactions containing private and shareable information across four domains. To our knowledge, MUMBench is the first public dataset designed to evaluate multiple memory operations, including retrieval, creation, update, and deletion, in a multi-user environment. Across three independent runs on MUMBench, AIM achieves 96.0% visibility classification accuracy, 58.8% strict operation accuracy, and 70.5% state-aware operation accuracy.

cs.AI↗

arXiv2026

Can We Trust LLM Judges: A Study of Capability-Dependent Biases and Multi-Judge Ensemble for Bias Calibration

LLMs are increasingly used as automated judges for model training and evaluation, yet individual judges exhibit systematic biases that undermine reliability. Much of prior work has studied biases in pairwise LLM-as-a-judge settings; in this paper, we focus on absolute scoring tasks, which mirror more realistic use cases. Across four benchmarks and six models (36 judge-examinee pairs), we show that a model's task accuracy strongly predicts its judging accuracy (Pearson $r \geq 0.90$ on most models) and inversely predicts its directional bias ($r \leq -0.83$), but that accuracy alone does not ensure fair evaluation: more capable examinee models consistently receive more lenient judgments from all judges ($r \geq 0.83$). To address this, we propose calibrated weighted majority voting (WMV), an ensemble evaluation method that aggregates multiple LLM judges weighted by online estimates of their false-positive and false-negative rates. We introduce a disagreement-based estimator that derives these error rates purely from inter-judge agreement patterns, requiring no ground-truth labels or task metadata. In a simulated experiment with shifting task distributions, our label-free WMV tracks an oracle with perfect error-rate knowledge to within 0.5 percentage points on average, outperforming both individual judges and unweighted majority voting. These results demonstrate that principled multi-judge calibration can simultaneously improve accuracy and correct for systematic leniency without requiring labeled data, offering a scalable path to reliable automated evaluation as model capabilities increase.

cs.LG↗

arXiv2026

Metag: A dataset to build agentic meta-reviewing capabilities

AI tools increasingly support tasks across the scientific research cycle, from experiment design and manuscript preparation to peer review. At the same time, the continuing growth in conference submissions has increased the burden on meta-reviewers, who must synthesize reviewer feedback, author rebuttals, and manuscript revisions. To address this concern, this paper introduces Metag, a dataset to accelerate the development of meta-reviewing agents, specifically to identify changes made to scientific articles during the review-rebuttal process. Each instance contains a reviewer concern, the author's proposed resolution, and the manuscript diffs implementing the stated change. Metag is collected by obtaining manuscript versions from before the review deadline and after acceptance, computing differences between the two documents, and asking human annotators to align these differences with action items from OpenReview discussions. The resulting dataset consists of 349 high-quality action items tied to paper differences and will enable building methods to empower meta reviewers to quickly identify whether authors have addressed reviewer statements and where in the paper those changes have been made, resulting in additional transparency and traceability throughout peer review. The dataset is publicly available at https://github.com/microsoft/Metag-dataset.

cs.LG↗

arXiv2026

CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments

Long-horizon reasoning is a key challenge for autonomous agents, yet existing benchmarks evaluate agents on single tasks in isolation. Real organizational work requires managing many concurrent long-horizon tasks with interleaving, dependencies, and reprioritization. We introduce Multi-Horizon Task Environments (MHTEs): a distinct problem class requiring coherent execution across dozens of interleaved tasks (45+, 500-1500+ steps) within persistent execution contexts spanning hours. We identify four failure modes that cause baseline CUAs to degrade from 16.7% to 8.7% completion as load scales 25% to 100%, a pattern consistent across three independent implementations. These failure modes are context saturation (O(N) vs O(1) growth), memory interference, dependency complexity (DAGs vs. chains), and reprioritization overhead. We present CorpGen, an architecture-agnostic framework addressing these failures via hierarchical planning for multi-horizon goal alignment, sub-agent isolation preventing cross-task contamination, tiered memory (working, structured, semantic), and adaptive summarization. CorpGen simulates corporate environments through digital employees with persistent identities and realistic schedules. Across three CUA backends (UFO2, OpenAI CUA, hierarchical) on OSWorld Office, CorpGen achieves up to 3.5x improvement over baselines (15.2% vs 4.3%) with stable performance under increasing load, confirming that gains stem from architectural mechanisms rather than specific CUA implementations. Ablation studies show experiential learning provides the largest gains.

cs.AI↗

arXiv2026

Fine-tuning Small Language Models as Efficient Enterprise Search Relevance Labelers

In enterprise search, building high-quality datasets at scale remains a central challenge due to the difficulty of acquiring labeled data. To resolve this challenge, we propose an efficient approach to fine-tune small language models (SLMs) for accurate relevance labeling, enabling high-throughput, domain-specific labeling comparable or even better in quality to that of state-of-the-art large language models (LLMs). To overcome the lack of high-quality and accessible datasets in the enterprise domain, our method leverages on synthetic data generation. Specifically, we employ an LLM to synthesize realistic enterprise queries from a seed document, apply BM25 to retrieve hard negatives, and use a teacher LLM to assign relevance scores. The resulting dataset is then distilled into an SLM, producing a compact relevance labeler. We evaluate our approach on a high-quality benchmark consisting of 923 enterprise query-document pairs annotated by trained human annotators, and show that the distilled SLM achieves agreement with human judgments on par with or better than the teacher LLM. Furthermore, our fine-tuned labeler substantially improves throughput, achieving 17 times increase while also being 19 times more cost-effective. This approach enables scalable and cost-effective relevance labeling for enterprise-scale retrieval applications, supporting rapid offline evaluation and iteration in real-world settings.

cs.IR↗

arXiv2025

High-Fidelity Synthetic ECG Generation via Mel-Spectrogram Informed Diffusion Training

The development of machine learning for cardiac care is severely hampered by privacy restrictions on sharing real patient electrocardiogram (ECG) data. Although generative AI offers a promising solution, the real-world use of existing model-synthesized ECGs is limited by persistent gaps in trustworthiness and clinical utility. In this work, we address two major shortcomings of current generative ECG methods: insufficient morphological fidelity and the inability to generate personalized, patient-specific physiological signals. To address these gaps, we build on a conditional diffusion-based Structured State Space Model (SSSD-ECG) with two principled innovations: (1) MIDT-ECG (Mel-Spectrogram Informed Diffusion Training), a novel training paradigm with time-frequency domain supervision to enforce physiological structural realism, and (2) multi-modal demographic conditioning to enable patient-specific synthesis. We comprehensively evaluate our approach on the PTB-XL dataset, assessing the synthesized ECG signals on fidelity, clinical coherence, privacy preservation, and downstream task utility. MIDT-ECG achieves substantial gains: it improves morphological coherence, preserves strong privacy guarantees with all metrics evaluated exceeding the baseline by 4-8%, and notably reduces the interlead correlation error by an average of 74%, while demographic conditioning enhances signal-to-noise ratio and personalization. In critical low-data regimes, a classifier trained on datasets supplemented with our synthetic ECGs achieves performance comparable to a classifier trained solely on real data. Together, we demonstrate that ECG synthesizers, trained with the proposed time-frequency structural regularization scheme, can serve as personalized, high-fidelity, privacy-preserving surrogates when real data are scarce, advancing the responsible use of generative AI in healthcare.

cs.LG↗

arXiv2025

Learning ECG Representations via Poly-Window Contrastive Learning

Electrocardiogram (ECG) analysis is foundational for cardiovascular disease diagnosis, yet the performance of deep learning models is often constrained by limited access to annotated data. Self-supervised contrastive learning has emerged as a powerful approach for learning robust ECG representations from unlabeled signals. However, most existing methods generate only pairwise augmented views and fail to leverage the rich temporal structure of ECG recordings. In this work, we present a poly-window contrastive learning framework. We extract multiple temporal windows from each ECG instance to construct positive pairs and maximize their agreement via statistics. Inspired by the principle of slow feature analysis, our approach explicitly encourages the model to learn temporally invariant and physiologically meaningful features that persist across time. We validate our approach through extensive experiments and ablation studies on the PTB-XL dataset. Our results demonstrate that poly-window contrastive learning consistently outperforms conventional two-view methods in multi-label superclass classification, achieving higher AUROC (0.891 vs. 0.888) and F1 scores (0.680 vs. 0.679) while requiring up to four times fewer pre-training epochs (32 vs. 128) and 14.8% in total wall clock pre-training time reduction. Despite processing multiple windows per sample, we achieve a significant reduction in the number of training epochs and total computation time, making our method practical for training foundational models. Through extensive ablations, we identify optimal design choices and demonstrate robustness across various hyperparameters. These findings establish poly-window contrastive learning as a highly efficient and scalable paradigm for automated ECG analysis and provide a promising general framework for self-supervised representation learning in biomedical time-series data.

cs.LG↗

arXiv2025

Benchmark Dataset Generation and Evaluation for Excel Formula Repair with LLMs

Excel is a pervasive yet often complex tool, particularly for novice users, where runtime errors arising from logical mistakes or misinterpretations of functions pose a significant challenge. While large language models (LLMs) offer promising assistance by explaining formula errors, the automated correction of these semantic runtime errors remains an open problem. A primary challenge to advancing models for such scenarios is the severe lack of high-quality, comprehensive datasets for training and rigorous evaluation. This paper addresses this gap by introducing a novel approach for constructing a benchmark dataset specifically designed for Excel formula repair. We propose a data generation pipeline, which leverages a small set of curated seed samples from online forums to synthetically expand the dataset. Our pipeline integrates few-shot prompting with LLMs and employs a robust \textit{LLM-as-a-Judge} validation framework, combined with execution-based checks to ensure the correctness and semantic fidelity of the generated data. This process produced a benchmark dataset of 618 high-quality samples, covering common runtime errors. Furthermore, we propose a context-aware baseline technique for Excel formula repair that utilizes LLMs to leverage both the faulty formula, and relevant spreadsheet context. We evaluate the performance of various LLMs (GPT-4o, GPT-4.1, Phi-3, Mistral) on our newly generated benchmark using execution-based metrics. Our analysis demonstrates the dataset's quality through manual annotation and provides insights into error and function distributions. The proposed generation methodology is highly scalable and can be readily adapted to create evaluation benchmarks for similar code repair tasks in other low-resource programming languages.

cs.SE↗

arXiv2025

Prostate-Specific Foundation Models for Enhanced Detection of Clinically Significant Cancer

Accurate prostate cancer diagnosis remains challenging. Even when using MRI, radiologists exhibit low specificity and significant inter-observer variability, leading to potential delays or inaccuracies in identifying clinically significant cancers. This leads to numerous unnecessary biopsies and risks of missing clinically significant cancers. Here we present prostate vision contrastive network (ProViCNet), prostate organ-specific vision foundation models for Magnetic Resonance Imaging (MRI) and Trans-Rectal Ultrasound imaging (TRUS) for comprehensive cancer detection. ProViCNet was trained and validated using 4,401 patients across six institutions, as a prostate cancer detection model on radiology images relying on patch-level contrastive learning guided by biopsy confirmed radiologist annotations. ProViCNet demonstrated consistent performance across multiple internal and external validation cohorts with area under the receiver operating curve values ranging from 0.875 to 0.966, significantly outperforming radiologists in the reader study (0.907 versus 0.805, p<0.001) for mpMRI, while achieving 0.670 to 0.740 for TRUS. We also integrated ProViCNet with standard PSA to develop a virtual screening test, and we showed that we can maintain the high sensitivity for detecting clinically significant cancers while more than doubling specificity from 15% to 38% (p<0.001), thereby substantially reducing unnecessary biopsies. These findings highlight that ProViCNet's potential for enhancing prostate cancer diagnosis accuracy and reduce unnecessary biopsies, thereby optimizing diagnostic pathways.

eess.IV↗

arXiv2025

Registration-Enhanced Segmentation Method for Prostate Cancer in Ultrasound Images

Prostate cancer is a major cause of cancer-related deaths in men, where early detection greatly improves survival rates. Although MRI-TRUS fusion biopsy offers superior accuracy by combining MRI's detailed visualization with TRUS's real-time guidance, it is a complex and time-intensive procedure that relies heavily on manual annotations, leading to potential errors. To address these challenges, we propose a fully automatic MRI-TRUS fusion-based segmentation method that identifies prostate tumors directly in TRUS images without requiring manual annotations. Unlike traditional multimodal fusion approaches that rely on naive data concatenation, our method integrates a registration-segmentation framework to align and leverage spatial information between MRI and TRUS modalities. This alignment enhances segmentation accuracy and reduces reliance on manual effort. Our approach was validated on a dataset of 1,747 patients from Stanford Hospital, achieving an average Dice coefficient of 0.212, outperforming TRUS-only (0.117) and naive MRI-TRUS fusion (0.132) methods, with significant improvements (p $<$ 0.01). This framework demonstrates the potential for reducing the complexity of prostate cancer diagnosis and provides a flexible architecture applicable to other multimodal medical imaging tasks.

eess.IV↗

arXiv2025

Multimodal MRI-Ultrasound AI for Prostate Cancer Detection Outperforms Radiologist MRI Interpretation: A Multi-Center Study

Pre-biopsy magnetic resonance imaging (MRI) is increasingly used to target suspicious prostate lesions. This has led to artificial intelligence (AI) applications improving MRI-based detection of clinically significant prostate cancer (CsPCa). However, MRI-detected lesions must still be mapped to transrectal ultrasound (TRUS) images during biopsy, which results in missing CsPCa. This study systematically evaluates a multimodal AI framework integrating MRI and TRUS image sequences to enhance CsPCa identification. The study included 3110 patients from three cohorts across two institutions who underwent prostate biopsy. The proposed framework, based on the 3D UNet architecture, was evaluated on 1700 test cases, comparing performance to unimodal AI models that use either MRI or TRUS alone. Additionally, the proposed model was compared to radiologists in a cohort of 110 patients. The multimodal AI approach achieved superior sensitivity (80%) and Lesion Dice (42%) compared to unimodal MRI (73%, 30%) and TRUS models (49%, 27%). Compared to radiologists, the multimodal model showed higher specificity (88% vs. 78%) and Lesion Dice (38% vs. 33%), with equivalent sensitivity (79%). Our findings demonstrate the potential of multimodal AI to improve CsPCa lesion targeting during biopsy and treatment planning, surpassing current unimodal models and radiologists; ultimately improving outcomes for prostate cancer patients.

eess.IV↗

arXiv2024

Mask Enhanced Deeply Supervised Prostate Cancer Detection on B-mode Micro-Ultrasound

Prostate cancer is a leading cause of cancer-related deaths among men. The recent development of high frequency, micro-ultrasound imaging offers improved resolution compared to conventional ultrasound and potentially a better ability to differentiate clinically significant cancer from normal tissue. However, the features of prostate cancer remain subtle, with ambiguous borders with normal tissue and large variations in appearance, making it challenging for both machine learning and humans to localize it on micro-ultrasound images. We propose a novel Mask Enhanced Deeply-supervised Micro-US network, termed MedMusNet, to automatically and more accurately segment prostate cancer to be used as potential targets for biopsy procedures. MedMusNet leverages predicted masks of prostate cancer to enforce the learned features layer-wisely within the network, reducing the influence of noise and improving overall consistency across frames. MedMusNet successfully detected 76% of clinically significant cancer with a Dice Similarity Coefficient of 0.365, significantly outperforming the baseline Swin-M2F in specificity and accuracy (Wilcoxon test, Bonferroni correction, p-value<0.05). While the lesion-level and patient-level analyses showed improved performance compared to human experts and different baseline, the improvements did not reach statistical significance, likely on account of the small cohort. We have presented a novel approach to automatically detect and segment clinically significant prostate cancer on B-mode micro-ultrasound images. Our MedMusNet model outperformed other models, surpassing even human experts. These preliminary results suggest the potential for aiding urologists in prostate cancer diagnosis via biopsy and treatment decision-making.

eess.IV↗

arXiv2024

Sound event detection based on auxiliary decoder and maximum probability aggregation for DCASE Challenge 2024 Task 4

In this report, we propose three novel methods for developing a sound event detection (SED) model for the DCASE 2024 Challenge Task 4. First, we propose an auxiliary decoder attached to the final convolutional block to improve feature extraction capabilities while reducing dependency on embeddings from pre-trained large models. The proposed auxiliary decoder operates independently from the main decoder, enhancing performance of the convolutional block during the initial training stages by assigning a different weight strategy between main and auxiliary decoder losses. Next, to address the time interval issue between the DESED and MAESTRO datasets, we propose maximum probability aggregation (MPA) during the training step. The proposed MPA method enables the model's output to be aligned with soft labels of 1 s in the MAESTRO dataset. Finally, we propose a multi-channel input feature that employs various versions of logmel and MFCC features to generate time-frequency pattern. The experimental results demonstrate the efficacy of these proposed methods in a view of improving SED performance by achieving a balanced enhancement across different datasets and label types. Ultimately, this approach presents a significant step forward in developing more robust and flexible SED models

eess.AS↗

arXiv2023

ProsDectNet: Bridging the Gap in Prostate Cancer Detection via Transrectal B-mode Ultrasound Imaging

Interpreting traditional B-mode ultrasound images can be challenging due to image artifacts (e.g., shadowing, speckle), leading to low sensitivity and limited diagnostic accuracy. While Magnetic Resonance Imaging (MRI) has been proposed as a solution, it is expensive and not widely available. Furthermore, most biopsies are guided by Transrectal Ultrasound (TRUS) alone and can miss up to 52% cancers, highlighting the need for improved targeting. To address this issue, we propose ProsDectNet, a multi-task deep learning approach that localizes prostate cancer on B-mode ultrasound. Our model is pre-trained using radiologist-labeled data and fine-tuned using biopsy-confirmed labels. ProsDectNet includes a lesion detection and patch classification head, with uncertainty minimization using entropy to improve model performance and reduce false positive predictions. We trained and validated ProsDectNet using a cohort of 289 patients who underwent MRI-TRUS fusion targeted biopsy. We then tested our approach on a group of 41 patients and found that ProsDectNet outperformed the average expert clinician in detecting prostate cancer on B-mode ultrasound images, achieving a patient-level ROC-AUC of 82%, a sensitivity of 74%, and a specificity of 67%. Our results demonstrate that ProsDectNet has the potential to be used as a computer-aided diagnosis system to improve targeted biopsy and treatment planning.

eess.IV↗

arXiv2022

Learn2Reg: comprehensive multi-task medical image registration challenge, dataset and evaluation in the era of deep learning

Image registration is a fundamental medical image analysis task, and a wide variety of approaches have been proposed. However, only a few studies have comprehensively compared medical image registration approaches on a wide range of clinically relevant tasks. This limits the development of registration methods, the adoption of research advances into practice, and a fair benchmark across competing approaches. The Learn2Reg challenge addresses these limitations by providing a multi-task medical image registration data set for comprehensive characterisation of deformable registration algorithms. A continuous evaluation will be possible at https://learn2reg.grand-challenge.org. Learn2Reg covers a wide range of anatomies (brain, abdomen, and thorax), modalities (ultrasound, CT, MR), availability of annotations, as well as intra- and inter-patient registration evaluation. We established an easily accessible framework for training and validation of 3D registration methods, which enabled the compilation of results of over 65 individual method submissions from more than 20 unique teams. We used a complementary set of metrics, including robustness, accuracy, plausibility, and runtime, enabling unique insight into the current state-of-the-art of medical image registration. This paper describes datasets, tasks, evaluation methods and results of the challenge, as well as results of further analysis of transferability to new datasets, the importance of label supervision, and resulting bias. While no single approach worked best across all tasks, many methodological aspects could be identified that push the performance of medical image registration to new state-of-the-art performance. Furthermore, we demystified the common belief that conventional registration methods have to be much slower than deep-learning-based methods.

eess.IV↗

arXiv2022

Domain Generalization for Prostate Segmentation in Transrectal Ultrasound Images: A Multi-center Study

Prostate biopsy and image-guided treatment procedures are often performed under the guidance of ultrasound fused with magnetic resonance images (MRI). Accurate image fusion relies on accurate segmentation of the prostate on ultrasound images. Yet, the reduced signal-to-noise ratio and artifacts (e.g., speckle and shadowing) in ultrasound images limit the performance of automated prostate segmentation techniques and generalizing these methods to new image domains is inherently difficult. In this study, we address these challenges by introducing a novel 2.5D deep neural network for prostate segmentation on ultrasound images. Our approach addresses the limitations of transfer learning and finetuning methods (i.e., drop in performance on the original training data when the model weights are updated) by combining a supervised domain adaptation technique and a knowledge distillation loss. The knowledge distillation loss allows the preservation of previously learned knowledge and reduces the performance drop after model finetuning on new datasets. Furthermore, our approach relies on an attention module that considers model feature positioning information to improve the segmentation accuracy. We trained our model on 764 subjects from one institution and finetuned our model using only ten subjects from subsequent institutions. We analyzed the performance of our method on three large datasets encompassing 2067 subjects from three different institutions. Our method achieved an average Dice Similarity Coefficient (Dice) of $94.0\pm0.03$ and Hausdorff Distance (HD95) of 2.28 $mm$ in an independent set of subjects from the first institution. Moreover, our model generalized well in the studies from the other two institutions (Dice: $91.0\pm0.03$; HD95: 3.7$mm$ and Dice: $82.0\pm0.03$; HD95: 7.1 $mm$).

eess.IV↗

arXiv2022

ConFUDA: Contrastive Fewshot Unsupervised Domain Adaptation for Medical Image Segmentation

Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to an unlabeled target domain. Contrastive learning (CL) in the context of UDA can help to better separate classes in feature space. However, in image segmentation, the large memory footprint due to the computation of the pixel-wise contrastive loss makes it prohibitive to use. Furthermore, labeled target data is not easily available in medical imaging, and obtaining new samples is not economical. As a result, in this work, we tackle a more challenging UDA task when there are only a few (fewshot) or a single (oneshot) image available from the target domain. We apply a style transfer module to mitigate the scarcity of target samples. Then, to align the source and target features and tackle the memory issue of the traditional contrastive loss, we propose the centroid-based contrastive learning (CCL) and a centroid norm regularizer (CNR) to optimize the contrastive pairs in both direction and magnitude. In addition, we propose multi-partition centroid contrastive learning (MPCCL) to further reduce the variance in the target features. Fewshot evaluation on MS-CMRSeg dataset demonstrates that ConFUDA improves the segmentation performance by 0.34 of the Dice score on the target domain compared with the baseline, and 0.31 Dice score improvement in a more rigorous oneshot setting.

cs.CV↗

arXiv2022

Few-shot Unsupervised Domain Adaptation for Multi-modal Cardiac Image Segmentation

Unsupervised domain adaptation (UDA) methods intend to reduce the gap between source and target domains by using unlabeled target domain and labeled source domain data, however, in the medical domain, target domain data may not always be easily available, and acquiring new samples is generally time-consuming. This restricts the development of UDA methods for new domains. In this paper, we explore the potential of UDA in a more challenging while realistic scenario where only one unlabeled target patient sample is available. We call it Few-shot Unsupervised Domain adaptation (FUDA). We first generate target-style images from source images and explore diverse target styles from a single target patient with Random Adaptive Instance Normalization (RAIN). Then, a segmentation network is trained in a supervised manner with the generated target images. Our experiments demonstrate that FUDA improves the segmentation performance by 0.33 of Dice score on the target domain compared with the baseline, and it also gives 0.28 of Dice score improvement in a more rigorous one-shot setting. Our code is available at \url{https://github.com/MingxuanGu/Few-shot-UDA}.

eess.IV↗