Do MLLMs Understand Pointing? Benchmarking and Enhancing Referential Reasoning in Egocentric Vision

Egocentric AI agents, such as smart glasses, rely on pointing gestures to resolve referential ambiguities in natural language commands. However, despite advancements in Multimodal Large Language Models (MLLMs), current systems often fail to precisely ground the spatial semantics of pointing. Instead, they rely on spurious correlations with visual proximity or object saliency, a phenomenon we term "Referential Hallucination." To address this gap, we introduce EgoPoint-Bench, a comprehensive question-answering benchmark designed to evaluate and enhance multimodal pointing reasoning in egocentric views. Comprising over 11k high-fidelity simulated and real-world samples, the benchmark spans five evaluation dimensions and three levels of referential complexity. Extensive experiments demonstrate that while state-of-the-art proprietary and open-source models struggle with egocentric pointing, models fine-tuned on our synthetic data achieve significant performance gains and robust sim-to-real generalization. This work highlights the importance of spatially aware supervision and offers a scalable path toward precise egocentric AI assistants. Project page: https://guyyyug.github.io/EgoPoint-Bench/

Towards Scalable Web Accessibility Audit with MLLMs as Copilots

Ensuring web accessibility is crucial for advancing social welfare, justice, and equality in digital spaces, yet the vast majority of website user interfaces remain non-compliant, due in part to the resource-intensive and unscalable nature of current auditing practices. While WCAG-EM offers a structured methodology for site-wise conformance evaluation, it involves great human efforts and lacks practical support for execution at scale. In this work, we present an auditing framework, AAA, which operationalizes WCAG-EM through a human-AI partnership model. AAA is anchored by two key innovations: GRASP, a graph-based multimodal sampling method that ensures representative page coverage via learned embeddings of visual, textual, and relational cues; and MaC, a multimodal large language model-based copilot that supports auditors through cross-modal reasoning and intelligent assistance in high-effort tasks. Together, these components enable scalable, end-to-end web accessibility auditing, empowering human auditors with AI-enhanced assistance for real-world impact. We further contribute four novel datasets designed for benchmarking core stages of the audit pipeline. Extensive experiments demonstrate the effectiveness of our methods, providing insights that small-scale language models can serve as capable experts when fine-tuned.

Learning neural representations for X-ray ptychography reconstruction with unknown probes

X-ray ptychography provides exceptional nanoscale resolution and is widely applied in materials science, biology, and nanotechnology. However, its full potential is constrained by the critical challenge of accurately reconstructing images when the illuminating probe is unknown. Conventional iterative methods and deep learning approaches are often suboptimal, particularly under the low-signal conditions inherent to low-dose and high-speed experiments. These limitations compromise reconstruction fidelity and restrict the broader adoption of the technique. In this work, we introduce the Ptychographic Implicit Neural Representation (PtyINR), a self-supervised framework that simultaneously addresses the object and probe recovery problem. By parameterizing both as continuous neural representations, PtyINR performs end-to-end reconstruction directly from raw diffraction patterns without requiring any pre-characterization of the probe. Extensive evaluations demonstrate that PtyINR achieves superior reconstruction quality on both simulated and experimental data, with remarkable robustness under challenging low-signal conditions. Furthermore, PtyINR offers a generalizable, physics-informed framework for addressing probe-dependent inverse problems, making it applicable to a wide range of computational microscopy problems.

Single-shot X-ray ptychography as a structured illumination method

Single-shot ptychography is a quantitative phase imaging method wherein overlapping beams of light arranged in a grid pattern simultaneously illuminate a sample, allowing a full ptychographic dataset to be collected in a single shot. It is primarily used at optical wavelengths, but there is interest in using it for X-ray imaging. However, the constraints imposed by X-ray optics have limited the resolution achievable to date. In this work, we reinterpret single-shot ptychography as a structured illumination method by viewing the grid of beams as a single, highly structured illumination function. Pre-calibrating this illumination and reconstructing single-shot data using the randomized probe imaging algorithm allows us to account for the overlap and coherent interference between the diffraction arising from each beam. We achieve a resolution 3.5 times finer than the numerical aperture-based limit imposed by traditional algorithms for single-shot ptychography. We argue that this reconstruction method will work better for most single-shot ptychography experiments and discuss the implications for the design of future single-shot X-ray microscopes.