Beyond Retrieval

AI agents operating on user interfaces must understand how interfaces communicate state and feedback to act reliably. As a core communicative modality, animations are increasingly used in modern interfaces, serving critical functional purposes beyond mere aesthetics. Thus, understanding UI animation is essential for comprehensive interface interpretation. However, recent studies of Vision Language Models (VLMs) for UI understanding have focused primarily on static screenshots, leaving it unclear how well these models handle dynamic UI animations. To address this gap, we created AniMINT, a novel dataset of 300 densely annotated UI animation videos. We systematically evaluate state-of-the-art VLMs on UI animation understanding, including their abilities to perceive the animation effects, identify animation purposes, and interpret animation meaning. Our results show that VLMs can reliably detect primitive motion. However, their high-level animation interpretation remains inconsistent, with substantial gaps relative to human performance. Finally, we use Motion, Context, and Perceptual Cues (MCPC) to probe factors affecting VLM performance, revealing key bottlenecks and directions for future improvement.

Echocardiography is a widely used modality for cardiac assessment due to its non-invasive and cost-effective nature, but the sparse and heterogeneous spatiotemporal views of the heart pose distinct challenges. Existing masked autoencoder (MAE) approaches typically process images or short clips independently, failing to capture the inherent multi-view structure required for coherent cardiac representation. We introduce Latent Attention Masked Autoencoder (LAMAE), a foundation model architecture tailored to the multi-view nature of medical imaging. LAMAE augments the standard MAE with a latent attention module that enables information exchange across frames and views directly in latent space. This allows the model to aggregate variable-length sequences and distinct views, reconstructing a holistic representation of cardiac function from partial observations. We pretrain LAMAE on MIMIC-IV-ECHO, a large-scale, uncurated dataset reflecting real-world clinical variability. To the best of our knowledge, we present the first results for predicting ICD-10 codes from MIMIC-IV-ECHO videos. Furthermore, we empirically demonstrate that representations learned from adult data transfer effectively to pediatric cohorts despite substantial anatomical differences. These results provide evidence that incorporating structural priors, such as multi-view attention, yields significantly more robust and transferable representations.

The advent of large language models (LLMs) has sparked an evolution of autonomous web browsing agents: given a web browsing task and serialised user interface (UI) state, an LLM is expected to suggest input actions that incrementally solve the given task. The central challenge lies in serialising UI state for LLMs. Web agents have increasingly relied on grounded graphical UI (GUI) snapshots - screenshots augmented with visual cues - favoured for their modest input token footprint. Document object model (DOM) snapshots, serialised as HTML, represent a compelling alternative that leverages previously demonstrated HTML interpretation capabilities of LLMs. Their excessive token footprint, however, has precluded reliable deployment with web agents to date. We propose D2Snap, an algorithm to downsample the DOM, premised on preserving actionability and actionability-discriminating features. We evaluate D2Snap-downsampled DOM snapshots using a snapshot-variant web agent (GPT-4o) on a dataset sampled from Online-Mind2Web. Whilst 42% of raw DOM snapshots exceed the model context window (128 x 10^3 tokens), all D2Snap-downsampled DOM snapshots of our reference configuration fit, at a mean context utilisation of 16.5%. Against the 67% success rate of a grounded GUI snapshot baseline, our configuration attains 73% (+5.8%pt; 95% CI -13.6 to +26.0%pt; McNemar, p = 0.47), excluding a deficit (one-sided 95%) beyond 11%pt. Image input moreover appears to add little to snapshot utility; grounding text alone attains 62% (-5.8%pt; McNemar, p = 0.37).

Out-of-Distribution (OoD) detection aims to justify whether a given sample is from the training distribution of the classifier-under-protection, i.e., In-Distribution (InD), or from an unknown out distribution. Recent researches have leveraged Diffusion Models (DMs) for OoD detection due to their powerful distribution modeling capability. Given an input image, an InD-pretrained DM produces a corresponding InD-aligned counterpart, which serves as a generative reference for comparison. However, existing DM-based methods typically assess this underlying discrepancy through visual-level distances in the raw image space, which may be misaligned with the distributional discrepancy relevant to OoD detection. In this work, we investigate the fundamentals of discrepancy assessment in DM-based OoD detection, asking how the discrepancy between an input and its DM-generated counterpart should be formulated, and in which representation spaces and with which metrics it should be measured. To this end, we propose to assess the discrepancy in a classifier-relative manner by exploiting the representation spaces of the classifier-under-protection, whose training on InD data encodes rich task-relevant InD knowledge. In particular, we quantify two types of discrepancy: feature-level covariate discrepancy in deep feature representations and logit-level concept discrepancy in output logits, enabling effective differentiation between InD and OoD samples. Moreover, a subspace-based strategy is devised to refine representations of the DM generation to promote discrepancy assessment. Together, these designs form our novel detection framework, namely DDR. Extensive experiments on the challenging large-scale ImageNet-1K dataset demonstrate the superior detection performance of DDR over both DM-based and non-DM-based methods.

Recent advances in large language models (LLMs) have made natural language interfaces (NLIs) widely accessible for data exploration, yet analysts who have a broad analytical objective still face the challenge of decomposing it into effective step-by-step queries, especially over unfamiliar, multi-table relational databases. Rather than generating high-level analytical agendas, we investigate how to augment an NLI with semantic- and context-aware next-step query recommendations that act as analytical scaffolding for relational database exploration. Our approach goes beyond interestingness-only methods by jointly integrating semantic relevance, data interestingness, and context coherence to guide exploration toward coherent, topic-focused analyses and potentially insightful subsets. We evaluate QRec-NLI with NL2SQL benchmarking, LLM-enhanced description validation, agentic comparisons against interestingness-only and LLM-based prompting baselines, and a 12-participant user study. In the agentic comparison, QRec-NLI yields more topically relevant and locally coherent query sequences than both baselines. In the user study against the interestingness-only baseline, it receives stronger ratings for insight-generation support and decision support.