Deterministic Hallucination Detection in Medical VQA via Confidence-Evidence Bayesian Gain

Multimodal large language models (MLLMs) have shown strong potential for medical Visual Question Answering (VQA), yet they remain prone to hallucinations, defined as generating responses that contradict the input image, posing serious risks in clinical settings. Current hallucination detection methods, such as Semantic Entropy (SE) and Vision-Amplified Semantic Entropy (VASE), require 10 to 20 stochastic generations per sample together with an external natural language inference model for semantic clustering, making them computationally expensive and difficult to deploy in practice. We observe that hallucinated responses exhibit a distinctive signature directly in the model's own log-probabilities: inconsistent token-level confidence and weak sensitivity to visual evidence. Based on this observation, we propose Confidence-Evidence Bayesian Gain (CEBaG), a deterministic hallucination detection method that requires no stochastic sampling, no external models, and no task-specific hyperparameters. CEBaG combines two complementary signals: token-level predictive variance, which captures inconsistent confidence across response tokens, and evidence magnitude, which measures how much the image shifts per-token predictions relative to text-only inference. Evaluated across four medical MLLMs and three VQA benchmarks (16 experimental settings), CEBaG achieves the highest AUC in 13 of 16 settings and improves over VASE by 8 AUC points on average, while being fully deterministic and self-contained. The code will be made available upon acceptance.

MARCUS: An agentic, multimodal vision-language model for cardiac diagnosis and management

Cardiovascular disease remains the leading cause of global mortality, with progress hindered by human interpretation of complex cardiac tests. Current AI vision-language models are limited to single-modality inputs and are non-interactive. We present MARCUS (Multimodal Autonomous Reasoning and Chat for Ultrasound and Signals), an agentic vision-language system for end-to-end interpretation of electrocardiograms (ECGs), echocardiograms, and cardiac magnetic resonance imaging (CMR) independently and as multimodal input. MARCUS employs a hierarchical agentic architecture comprising modality-specific vision-language expert models, each integrating domain-trained visual encoders with multi-stage language model optimization, coordinated by a multimodal orchestrator. Trained on 13.5 million images (0.25M ECGs, 1.3M echocardiogram images, 12M CMR images) and our novel expert-curated dataset spanning 1.6 million questions, MARCUS achieves state-of-the-art performance surpassing frontier models (GPT-5 Thinking, Gemini 2.5 Pro Deep Think). Across internal (Stanford) and external (UCSF) test cohorts, MARCUS achieves accuracies of 87-91% for ECG, 67-86% for echocardiography, and 85-88% for CMR, outperforming frontier models by 34-45% (P<0.001). On multimodal cases, MARCUS achieved 70% accuracy, nearly triple that of frontier models (22-28%), with 1.7-3.0x higher free-text quality scores. Our agentic architecture also confers resistance to mirage reasoning, whereby vision-language models derive reasoning from unintended textual signals or hallucinated visual content. MARCUS demonstrates that domain-specific visual encoders with an agentic orchestrator enable multimodal cardiac interpretation. We release our models, code, and benchmark open-source.

Mirage The Illusion of Visual Understanding

Multimodal AI systems have achieved remarkable performance across a broad range of real-world tasks, yet the mechanisms underlying visual-language reasoning remain surprisingly poorly understood. We report three findings that challenge prevailing assumptions about how these systems process and integrate visual information. First, Frontier models readily generate detailed image descriptions and elaborate reasoning traces, including pathology-biased clinical findings, for images never provided; we term this phenomenon mirage reasoning. Second, without any image input, models also attain strikingly high scores across general and medical multimodal benchmarks, bringing into question their utility and design. In the most extreme case, our model achieved the top rank on a standard chest X-ray question-answering benchmark without access to any images. Third, when models were explicitly instructed to guess answers without image access, rather than being implicitly prompted to assume images were present, performance declined markedly. Explicit guessing appears to engage a more conservative response regime, in contrast to the mirage regime in which models behave as though images have been provided. These findings expose fundamental vulnerabilities in how visual-language models reason and are evaluated, pointing to an urgent need for private benchmarks that eliminate textual cues enabling non-visual inference, particularly in medical contexts where miscalibrated AI carries the greatest consequence. We introduce B-Clean as a principled solution for fair, vision-grounded evaluation of multimodal AI systems.

Interpretable Cross-Network Attention for Resting-State fMRI Representation Learning

Understanding how large-scale functional brain networks reorganize during cognitive decline remains a central challenge in neuroimaging. While recent self-supervised models have shown promise for learning representations from resting-state fMRI, their internal mechanisms are difficult to interpret, limiting mechanistic insight. We propose BrainInterNet, a network-aware self-supervised framework based on masked reconstruction with cross-attention that explicitly models inter-network dependencies in rs-fMRI. By selectively masking predefined functional networks and reconstructing them from remaining context, our approach enables direct quantification of network predictability and interpretable analysis of cross-network interactions. We train BrainInterNet on multi-cohort fMRI data (from the ABCD, HCP Development, HCP Young Adults, and HCP Aging datasets) and evaluate on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, in total comprising 5,582 recordings. Our method reveals systematic alterations in the brain's network interactions under AD, including in the default mode, limbic, and attention networks. In parallel, the learned representations support accurate Alzheimer's-spectrum classification and yield a compact summary marker that tracks disease severity longitudinally. Together, these results demonstrate that network-guided masked modeling with cross-attention provides an interpretable and effective framework for characterizing functional reorganization in neurodegeneration.

Latent Forcing: Reordering the Diffusion Trajectory for Pixel-Space Image Generation

Latent diffusion models excel at generating high-quality images but lose the benefits of end-to-end modeling. They discard information during image encoding, require a separately trained decoder, and model an auxiliary distribution to the raw data. In this paper, we propose Latent Forcing, a simple modification to existing architectures that achieves the efficiency of latent diffusion while operating on raw natural images. Our approach orders the denoising trajectory by jointly processing latents and pixels with separately tuned noise schedules. This allows the latents to act as a scratchpad for intermediate computation before high-frequency pixel features are generated. We find that the order of conditioning signals is critical, and we analyze this to explain differences between REPA distillation in the tokenizer and the diffusion model, conditional versus unconditional generation, and how tokenizer reconstruction quality relates to diffusability. Applied to ImageNet, Latent Forcing achieves a new state-of-the-art for diffusion transformer-based pixel generation at our compute scale.

Anatomically Guided Latent Diffusion for Brain MRI Progression Modeling

Accurately modeling longitudinal brain MRI progression is crucial for understanding neurodegenerative diseases and predicting individualized structural changes. Existing state-of-the-art approaches, such as Brain Latent Progression (BrLP), often use multi-stage training pipelines with auxiliary conditioning modules but suffer from architectural complexity, suboptimal use of conditional clinical covariates, and limited guarantees of anatomical consistency. We propose Anatomically Guided Latent Diffusion Model (AG-LDM), a segmentation-guided framework that enforces anatomically consistent progression while substantially simplifying the training pipeline. AG-LDM conditions latent diffusion by directly fusing baseline anatomy, noisy follow-up states, and clinical covariates at the input level, a strategy that avoids auxiliary control networks by learning a unified, end-to-end model that represents both anatomy and progression. A lightweight 3D tissue segmentation model (WarpSeg) provides explicit anatomical supervision during both autoencoder fine-tuning and diffusion model training, ensuring consistent brain tissue boundaries and morphometric fidelity. Experiments on 31,713 ADNI longitudinal pairs and zero-shot evaluation on OASIS-3 demonstrate that AG-LDM matches or surpasses more complex diffusion models, achieving state-of-the-art image quality and 15-20\% reduction in volumetric errors in generated images. AG-LDM also exhibits markedly stronger utilization of temporal and clinical covariates (up to 31.5x higher sensitivity than BrLP) and generates biologically plausible counterfactual trajectories, accurately capturing hallmarks of Alzheimer's progression such as limbic atrophy and ventricular expansion. These results highlight AG-LDM as an efficient, anatomically grounded framework for reliable brain MRI progression modeling.

VideoWeave: A Data-Centric Approach for Efficient Video Understanding

Training video-language models is often prohibitively expensive due to the high cost of processing long frame sequences and the limited availability of annotated long videos. We present VideoWeave, a simple yet effective approach to improve data efficiency by constructing synthetic long-context training samples that splice together short, captioned videos from existing datasets. Rather than modifying model architectures or optimization objectives, VideoWeave reorganizes available video-text pairs to expand temporal diversity within fixed compute. We systematically study how different data composition strategies like random versus visually clustered splicing and caption enrichment affect downstream performance on downstream video question answering. Under identical compute constraints, models trained with VideoWeave achieve higher accuracy than conventional video finetuning. Our results highlight that reorganizing training data, rather than altering architectures, may offer a simple and scalable path for training video-language models. We link our code for all experiments here.

A Tool Bottleneck Framework for Clinically-Informed and Interpretable Medical Image Understanding

Recent tool-use frameworks powered by vision-language models (VLMs) improve image understanding by grounding model predictions with specialized tools. Broadly, these frameworks leverage VLMs and a pre-specified toolbox to decompose the prediction task into multiple tool calls (often deep learning models) which are composed to make a prediction. The dominant approach to composing tools is using text, via function calls embedded in VLM-generated code or natural language. However, these methods often perform poorly on medical image understanding, where salient information is encoded as spatially-localized features that are difficult to compose or fuse via text alone. To address this, we propose a tool-use framework for medical image understanding called the Tool Bottleneck Framework (TBF), which composes VLM-selected tools using a learned Tool Bottleneck Model (TBM). For a given image and task, TBF leverages an off-the-shelf medical VLM to select tools from a toolbox that each extract clinically-relevant features. Instead of text-based composition, these tools are composed by the TBM, which computes and fuses the tool outputs using a neural network before outputting the final prediction. We propose a simple and effective strategy for TBMs to make predictions with any arbitrary VLM tool selection. Overall, our framework not only improves tool-use in medical imaging contexts, but also yields more interpretable, clinically-grounded predictors. We evaluate TBF on tasks in histopathology and dermatology and find that these advantages enable our framework to perform on par with or better than deep learning-based classifiers, VLMs, and state-of-the-art tool-use frameworks, with particular gains in data-limited regimes. Our code is available at https://github.com/christinaliu2020/tool-bottleneck-framework.

QuantiPhy: A Quantitative Benchmark Evaluating Physical Reasoning Abilities of Vision-Language Models

Understanding the physical world is essential for generalist AI agents. However, it remains unclear whether state-of-the-art vision perception models (e.g., large VLMs) can reason physical properties quantitatively. Existing evaluations are predominantly VQA-based and qualitative, offering limited insight into whether these models can infer the kinematic quantities of moving objects from video observations. To address this, we present QuantiPhy, the first benchmark designed to quantitatively measure a VLM's physical reasoning ability. Comprising more than 3.3K video-text instances with numerical ground truth, QuantiPhy evaluates a VLM's performance on estimating an object's size, velocity, and acceleration at a given timestamp, using one of these properties as an input prior. The benchmark standardizes prompts and scoring to assess numerical accuracy, enabling fair comparisons across models. Our experiments on state-of-the-art VLMs reveal a consistent gap between their qualitative plausibility and actual numerical correctness. We further provide an in-depth analysis of key factors like background noise, counterfactual priors, and strategic prompting and find that state-of-the-art VLMs lean heavily on pre-trained world knowledge rather than faithfully using the provided visual and textual inputs as references when reasoning kinematic properties quantitatively. QuantiPhy offers the first rigorous, scalable testbed to move VLMs beyond mere verbal plausibility toward a numerically grounded physical understanding.

Repurposing 2D Diffusion Models for 3D Shape Completion

We present a framework that adapts 2D diffusion models for 3D shape completion from incomplete point clouds. While text-to-image diffusion models have achieved remarkable success with abundant 2D data, 3D diffusion models lag due to the scarcity of high-quality 3D datasets and a persistent modality gap between 3D inputs and 2D latent spaces. To overcome these limitations, we introduce the Shape Atlas, a compact 2D representation of 3D geometry that (1) enables full utilization of the generative power of pretrained 2D diffusion models, and (2) aligns the modalities between the conditional input and output spaces, allowing more effective conditioning. This unified 2D formulation facilitates learning from limited 3D data and produces high-quality, detail-preserving shape completions. We validate the effectiveness of our results on the PCN and ShapeNet-55 datasets. Additionally, we show the downstream application of creating artist-created meshes from our completed point clouds, further demonstrating the practicality of our method.