Cross-Modal Memory Compression for Efficient Multi-Agent Debate

Multi-agent debate can improve reasoning quality and reduce hallucinations, but it incurs rapidly growing context as debate rounds and agent count increase. Retaining full textual histories leads to token usage that can exceed context limits and often requires repeated summarization, adding overhead and compounding information loss. We introduce DebateOCR, a cross-modal compression framework that replaces long textual debate traces with compact image representations, which are then consumed through a dedicated vision encoder to condition subsequent rounds. This design compresses histories that commonly span tens to hundreds of thousands of tokens, cutting input tokens by more than 92% and yielding substantially lower compute cost and faster inference across multiple benchmarks. We further provide a theoretical perspective showing that diversity across agents supports recovery of omitted information: although any single compressed history may discard details, aggregating multiple agents' compressed views allows the collective representation to approach the information bottleneck with exponentially high probability.

OS-Marathon: Benchmarking Computer-Use Agents on Long-Horizon Repetitive Tasks

Long-horizon, repetitive workflows are common in professional settings, such as processing expense reports from receipts and entering student grades from exam papers. These tasks are often tedious for humans since they can extend to extreme lengths proportional to the size of the data to process. However, they are ideal for Computer-Use Agents (CUAs) due to their structured, recurring sub-workflows with logic that can be systematically learned. Identifying the absence of an evaluation benchmark as a primary bottleneck, we establish OS-Marathon, comprising 242 long-horizon, repetitive tasks across 2 domains to evaluate state-of-the-art (SOTA) agents. We then introduce a cost-effective method to construct a condensed demonstration using only few-shot examples to teach agents the underlying workflow logic, enabling them to execute similar workflows effectively on larger, unseen data collections. Extensive experiments demonstrate both the inherent challenges of these tasks and the effectiveness of our proposed method. Project website: https://os-marathon.github.io/.

A multitask framework for automated interpretation of multi-frame right upper quadrant ultrasound in clinical decision support

Ultrasound is a cornerstone of emergency and hepatobiliary imaging, yet its interpretation remains highly operator-dependent and time-sensitive. Here, we present a multitask vision-language agent (VLM) developed to assist with comprehensive right upper quadrant (RUQ) ultrasound interpretation across the full diagnostic workflow. The system was trained on a large, multi-center dataset comprising a primary cohort from Johns Hopkins Medical Institutions (9,189 cases, 594,099 images) and externally validated on cohorts from Stanford University (108 cases, 3,240 images) and a major Chinese medical center (257 cases, 3,178 images). Built on the Qwen2.5-VL-7B architecture, the agent integrates frame-level visual understanding with report-grounded language reasoning to perform three tasks: (i) classification of 18 hepatobiliary and gallbladder conditions, (ii) generation of clinically coherent diagnostic reports, and (iii) surgical decision support based on ultrasound findings and clinical data. The model achieved high diagnostic accuracy across all tasks, generated reports that were indistinguishable from expert-written versions in blinded evaluations, and demonstrated superior factual accuracy and information density on content-based metrics. The agent further identified patients requiring cholecystectomy with high precision, supporting real-time decision-making. These results highlight the potential of generalist vision-language models to improve diagnostic consistency, reporting efficiency, and surgical triage in real-world ultrasound practice.

Internal Representations as Indicators of Hallucinations in Agent Tool Selection

Large Language Models (LLMs) have shown remarkable capabilities in tool calling and tool usage, but suffer from hallucinations where they choose incorrect tools, provide malformed parameters and exhibit 'tool bypass' behavior by performing simulations and generating outputs instead of invoking specialized tools or external systems. This undermines the reliability of LLM based agents in production systems as it leads to inconsistent results, and bypasses security and audit controls. Such hallucinations in agent tool selection require early detection and error handling. Unlike existing hallucination detection methods that require multiple forward passes or external validation, we present a computationally efficient framework that detects tool-calling hallucinations in real-time by leveraging LLMs' internal representations during the same forward pass used for generation. We evaluate this approach on reasoning tasks across multiple domains, demonstrating strong detection performance (up to 86.4\% accuracy) while maintaining real-time inference capabilities with minimal computational overhead, particularly excelling at detecting parameter-level hallucinations and inappropriate tool selections, critical for reliable agent deployment.

Beyond the LUMIR challenge: The pathway to foundational registration models

Medical image challenges have played a transformative role in advancing the field, catalyzing algorithmic innovation and establishing new performance standards across diverse clinical applications. Image registration, a foundational task in neuroimaging pipelines, has similarly benefited from the Learn2Reg initiative. Building on this foundation, we introduce the Large-scale Unsupervised Brain MRI Image Registration (LUMIR) challenge, a next-generation benchmark designed to assess and advance unsupervised brain MRI registration. Distinct from prior challenges that leveraged anatomical label maps for supervision, LUMIR removes this dependency by providing over 4,000 preprocessed T1-weighted brain MRIs for training without any label maps, encouraging biologically plausible deformation modeling through self-supervision. In addition to evaluating performance on 590 held-out test subjects, LUMIR introduces a rigorous suite of zero-shot generalization tasks, spanning out-of-domain imaging modalities (e.g., FLAIR, T2-weighted, T2*-weighted), disease populations (e.g., Alzheimer's disease), acquisition protocols (e.g., 9.4T MRI), and species (e.g., macaque brains). A total of 1,158 subjects and over 4,000 image pairs were included for evaluation. Performance was assessed using both segmentation-based metrics (Dice coefficient, 95th percentile Hausdorff distance) and landmark-based registration accuracy (target registration error). Across both in-domain and zero-shot tasks, deep learning-based methods consistently achieved state-of-the-art accuracy while producing anatomically plausible deformation fields. The top-performing deep learning-based models demonstrated diffeomorphic properties and inverse consistency, outperforming several leading optimization-based methods, and showing strong robustness to most domain shifts, the exception being a drop in performance on out-of-domain contrasts.

Optimizing Prompt Strategies for SAM: Advancing lesion Segmentation Across Diverse Medical Imaging Modalities

Purpose: To evaluate various Segmental Anything Model (SAM) prompt strategies across four lesions datasets and to subsequently develop a reinforcement learning (RL) agent to optimize SAM prompt placement. Materials and Methods: This retrospective study included patients with four independent ovarian, lung, renal, and breast tumor datasets. Manual segmentation and SAM-assisted segmentation were performed for all lesions. A RL model was developed to predict and select SAM points to maximize segmentation performance. Statistical analysis of segmentation was conducted using pairwise t-tests. Results: Results show that increasing the number of prompt points significantly improves segmentation accuracy, with Dice coefficients rising from 0.272 for a single point to 0.806 for five or more points in ovarian tumors. The prompt location also influenced performance, with surface and union-based prompts outperforming center-based prompts, achieving mean Dice coefficients of 0.604 and 0.724 for ovarian and breast tumors, respectively. The RL agent achieved a peak Dice coefficient of 0.595 for ovarian tumors, outperforming random and alternative RL strategies. Additionally, it significantly reduced segmentation time, achieving a nearly 10-fold improvement compared to manual methods using SAM. Conclusion: While increased SAM prompts and non-centered prompts generally improved segmentation accuracy, each pathology and modality has specific optimal thresholds and placement strategies. Our RL agent achieved superior performance compared to other agents while achieving a significant reduction in segmentation time.

Optimizing prompt strategies for the Segment Anything Model are explored, focusing on prompt location, number, and reinforcement learning-based agent for prompt placement across four lesion datasets

Purpose: To evaluate various Segmental Anything Model (SAM) prompt strategies across four lesions datasets and to subsequently develop a reinforcement learning (RL) agent to optimize SAM prompt placement. Materials and Methods: This retrospective study included patients with four independent ovarian, lung, renal, and breast tumor datasets. Manual segmentation and SAM-assisted segmentation were performed for all lesions. A RL model was developed to predict and select SAM points to maximize segmentation performance. Statistical analysis of segmentation was conducted using pairwise t-tests. Results: Results show that increasing the number of prompt points significantly improves segmentation accuracy, with Dice coefficients rising from 0.272 for a single point to 0.806 for five or more points in ovarian tumors. The prompt location also influenced performance, with surface and union-based prompts outperforming center-based prompts, achieving mean Dice coefficients of 0.604 and 0.724 for ovarian and breast tumors, respectively. The RL agent achieved a peak Dice coefficient of 0.595 for ovarian tumors, outperforming random and alternative RL strategies. Additionally, it significantly reduced segmentation time, achieving a nearly 10-fold improvement compared to manual methods using SAM. Conclusion: While increased SAM prompts and non-centered prompts generally improved segmentation accuracy, each pathology and modality has specific optimal thresholds and placement strategies. Our RL agent achieved superior performance compared to other agents while achieving a significant reduction in segmentation time.

MUNBa: Machine Unlearning via Nash Bargaining

Machine Unlearning (MU) aims to selectively erase harmful behaviors from models while retaining the overall utility of the model. As a multi-task learning problem, MU involves balancing objectives related to forgetting specific concepts/data and preserving general performance. A naive integration of these forgetting and preserving objectives can lead to gradient conflicts, impeding MU algorithms from reaching optimal solutions. To address the gradient conflict issue, we reformulate MU as a two-player cooperative game, where the two players, namely, the forgetting player and the preservation player, contribute via their gradient proposals to maximize their overall gain. To this end, inspired by the Nash bargaining theory, we derive a closed-form solution to guide the model toward the Pareto front, effectively avoiding the gradient conflicts. Our formulation of MU guarantees an equilibrium solution, where any deviation from the final state would lead to a reduction in the overall objectives for both players, ensuring optimality in each objective. We evaluate our algorithm's effectiveness on a diverse set of tasks across image classification and image generation. Extensive experiments with ResNet, vision-language model CLIP, and text-to-image diffusion models demonstrate that our method outperforms state-of-the-art MU algorithms, achieving superior performance on several benchmarks. For example, in the challenging scenario of sample-wise forgetting, our algorithm approaches the gold standard retrain baseline. Our results also highlight improvements in forgetting precision, preservation of generalization, and robustness against adversarial attacks.

TabDeco: A Comprehensive Contrastive Framework for Decoupled Representations in Tabular Data

Representation learning is a fundamental aspect of modern artificial intelligence, driving substantial improvements across diverse applications. While selfsupervised contrastive learning has led to significant advancements in fields like computer vision and natural language processing, its adaptation to tabular data presents unique challenges. Traditional approaches often prioritize optimizing model architecture and loss functions but may overlook the crucial task of constructing meaningful positive and negative sample pairs from various perspectives like feature interactions, instance-level patterns and batch-specific contexts. To address these challenges, we introduce TabDeco, a novel method that leverages attention-based encoding strategies across both rows and columns and employs contrastive learning framework to effectively disentangle feature representations at multiple levels, including features, instances and data batches. With the innovative feature decoupling hierarchies, TabDeco consistently surpasses existing deep learning methods and leading gradient boosting algorithms, including XG-Boost, CatBoost, and LightGBM, across various benchmark tasks, underscoring its effectiveness in advancing tabular data representation learning.

DSCformer: A Dual-Branch Network Integrating Enhanced Dynamic Snake Convolution and SegFormer for Crack Segmentation

In construction quality monitoring, accurately detecting and segmenting cracks in concrete structures is paramount for safety and maintenance. Current convolutional neural networks (CNNs) have demonstrated strong performance in crack segmentation tasks, yet they often struggle with complex backgrounds and fail to capture fine-grained tubular structures fully. In contrast, Transformers excel at capturing global context but lack precision in detailed feature extraction. We introduce DSCformer, a novel hybrid model that integrates an enhanced Dynamic Snake Convolution (DSConv) with a Transformer architecture for crack segmentation to address these challenges. Our key contributions include the enhanced DSConv through a pyramid kernel for adaptive offset computation and a simultaneous bi-directional learnable offset iteration, significantly improving the model's performance to capture intricate crack patterns. Additionally, we propose a Weighted Convolutional Attention Module (WCAM), which refines channel attention, allowing for more precise and adaptive feature attention. We evaluate DSCformer on the Crack3238 and FIND datasets, achieving IoUs of 59.22\% and 87.24\%, respectively. The experimental results suggest that our DSCformer outperforms state-of-the-art methods across different datasets.