Author Correction: A broadly protective antibody targeting gammaherpesvirus gB
该条目无摘要(多为勘误、社论或新闻类内容,出版方未提供摘要)
Academic Intelligence · Curated Daily
AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。
该条目无摘要(多为勘误、社论或新闻类内容,出版方未提供摘要)
Spatiotemporal intelligence in autonomous driving (AD) requires an agent to integrate multi-view observations into a coherent scene representation, maintain object continuity across viewpoints and time, and reason about spatial relations, interactions, and future dynamics. However, existing AD vision-language benchmarks largely focus on single-view, static, ego-centric, or single-source question answering, leaving it unclear whether current Vision-Language Models (VLMs) can truly construct and reason over dynamic driving scenes. We introduce DriveSpatial, a benchmark of 15.6K human-verified QA pairs across 20 tasks from five large-scale AD datasets. DriveSpatial evaluates four abilities: Cognitive Scene Construction, Multi-view Relational Understanding, Temporal Reasoning, and Generalization. Unlike prior benchmarks, DriveSpatial is generated from a dynamic multi-relational scene graph that encodes object states, spatial relations, interactions, camera visibility, and temporal correspondences, enabling QA pairs that enforce genuine cross-view and spatiotemporal reasoning. Evaluating 15 representative VLMs reveals a substantial human-model gap: the strongest model trails humans by 28.4 points, with Cognitive Scene Construction emerging as the key bottleneck. Further diagnostics show that language-only prompting is insufficient, while explicit BEV grounding consistently improves performance. These results suggest that current VLMs lack the scene-construction ability needed for reliable spatiotemporal driving intelligence. DriveSpatial and its construction pipeline will be released to support future research.
arXiv:2511.05221v3 Announce Type: replace Abstract: Isolated rapid eye movement sleep behavior disorder (iRBD) is a major prodromal marker of $\alpha$-synucleinopathies, often preceding the clinical onset of Parkinson's disease, dementia with Lewy bodies, or multiple system atrophy. While wrist-worn actimeters hold significant potential for detecting RBD in large-scale screening efforts by capturing abnormal nocturnal movements, they become inoperable without a reliable and efficient analysis pipeline. This study presents ActiTect, a fully automated, open-source machine learning tool to identify RBD from actigraphy recordings. To ensure generalizability across heterogeneous acquisition settings, our pipeline includes robust preprocessing and automated sleep-wake detection to harmonize multi-device data and extract physiologically interpretable motion features characterizing activity patterns. Model development was conducted on a cohort of 78 individuals, yielding strong discrimination under nested cross-validation (AUROC = 0.95). Generalization was confirmed on a blinded local test set (n = 31, AUROC = 0.86) and on two independent external cohorts (n = 113, AUROC = 0.84; n = 57, AUROC = 0.94). To assess real-world robustness, leave-one-dataset-out cross-validation across the internal and external cohorts demonstrated consistent performance (AUROC range = 0.84-0.89). A complementary stability analysis showed that key predictive features remained reproducible across datasets, supporting the final pooled multi-center model as a robust pre-trained resource for broader deployment. By being open-source and easy to use, our tool promotes widespread adoption and facilitates independent validation and collaborative improvements, thereby advancing the field toward a unified and generalizable RBD detection model using wearable devices.
arXiv:2606.18469v1 Announce Type: cross Abstract: Neuroscientific research has revealed that the brain encodes complex behaviors by leveraging structured, low-dimensional manifolds and dynamically fusing multiple sources of information through adaptive gating mechanisms. Inspired by these principles, we propose a novel reinforcement learning (RL) framework that encourages the disentanglement of dynamics-specific and reward-specific features, drawing direct parallels to how neural circuits separate and integrate information for efficient decision-making. Our approach leverages locally linear embeddings (LLEs) to capture the intrinsic, locally linear structure inherent in many environments, mirroring the local smoothness observed in neural population activity, while concurrently deriving reward-specific features through the standard RL objective. An attention mechanism, analogous to cortical gating, adaptively fuses these complementary representations on a per-state basis. Experimental results on benchmark tasks demonstrate that our method, grounded in neuroscientific principles, improves learning efficiency and overall performance compared to conventional RL approaches, highlighting the benefits of explicitly modeling local state structures and adaptive feature selection as observed in biological systems.
arXiv:2605.18784v2 Announce Type: replace-cross Abstract: The rapid diffusion of agentic AI has created a new coverage problem for commercial insurance: some AI-mediated losses are now affirmatively insured, some create silent-AI exposure under legacy cyber, technology errors-and-omissions (E&O), directors-and-officers (D&O), employment practices liability (EPLI), crime, and media policies, and others are being actively excluded. This paper maps that emerging boundary by coding 55 AI threat classes against 26 insurance products, endorsements, and exclusion regimes using public carrier materials and OWASP/MITRE threat catalogs. We identify a four-tier insurability frontier: affirmatively insured perils, silent-AI exposures, actively excluded perils, and perils outside conventional private insurance structures. Our coding measures publicly claimed positioning rather than executed contract wording; the headline statistics describe what carriers publicly state about coverage, not what would be paid in any specific claim. Three patterns emerge. First, affirmative AI coverage is beginning to differentiate by primary risk emphasis: public materials often position Munich Re around model performance and drift, Armilla and parts of the Lloyd's market around hallucination and broader AI liability, Tokio Marine Kiln and CFC around IP and technology E&O concerns, Apollo ibott around emerging autonomous system liability, and Coalition around deepfake and AI-enabled cyber response. Second, legacy lines retain silent-AI exposure where AI is an instrumentality rather than the legal cause of loss. Third, foundation model concentration is the clearest genuinely novel insurability frontier because upstream model failure can correlate losses across many cedents at once; the relevant market design question is which insurability constraint each candidate structure relaxes, not merely which systemic risk template exists.
Conversational shopping assistants now serve hundreds of millions of customers, yet no existing benchmark jointly evaluates the open-ended multi-turn reasoning, domain expertise, and criterion-level quality that real shopping conversations demand. Shopping reasoning is unique among language model applications. Unlike factual question answering or verifiable code generation, it requires balancing subjective preferences, budget constraints, and cross-product trade-offs across multi-turn dialogue, capabilities absent from previous e-commerce and general-purpose benchmarks. We introduce the Shopping Reasoning Bench, an expert-authored benchmark of 525 missions (232 single-turn, 293 multi-turn) with 10863 importance-weighted binary rubrics authored by retail domain experts. These criteria are organized under a taxonomy of five reasoning categories and fifteen subcategories covering diverse demands such as preference refinement, trade-off analysis, and compatibility assessment. An evaluation of nine models across three families (GPT, Claude, Gemini) shows that pass rates reach only 57–77% overall. On multi-turn missions, all models score 13–29 points lower on optional above-and-beyond criteria than on required ones, and performance degrades 4–18 points as conversations progress. These gaps show that current models handle basic shopping assistance but fall short of expert-level advice, making Shopping Reasoning Bench a challenging testbed for future shopping assistant development.
arXiv:2606.19207v1 Announce Type: new Abstract: We introduce a family of Gaussian fields whose covariance structure exhibits an inhomogeneous, microscopic slowdown and it interpolates between a $\log$ profile (for a certain interpolation parameter $\alpha=0$) and a $\log\log$ profile (when the interpolation parameter is $\alpha=1/2$). We consider both one dimensional such objects (which we call {\it Branching Brownian Motions in a cooling environment}) as well as higher dimensional, spatial fields. We identify the correct centering of the maximum at time $T$ and prove tightness of the recentered maximum. While the exponent in the first-order growth varies linearly with $\alpha$, giving a leading order of $T^{1-\alpha}$, the second-order correction exhibits a phase transition at $\alpha=1/3$.
Spoken dialogue models typically start from text LLM backbones, yet reasoning often degrades when conditioning on speech instead of text. We attribute part of this modality gap to a temporal-granularity mismatch: speech tokens are temporally redundant and far longer than text under matched semantics, diluting per-token semantic density and weakening text-native reasoning dynamics. We study speech token design as a representation selection problem and sweep frame rates under a frozen LLM backbone with a fixed information rate. To make low frame rates feasible, we introduce factorized FSQ and a lightweight non-autoregressive audio LM head, scaling capacity to nearly 300\,bits/frame without sacrificing efficient prediction. With the bottleneck removed, we sweep frame rates (50$\rightarrow$2.08\,Hz) and alignment depth, and observe a consistent best regime for speech QA at 4.17\,Hz with intermediate-layer representation alignment.
arXiv:2606.11092v2 Announce Type: replace-cross Abstract: Elite humanoid soccer shooting requires whole-body stability, high-impulse whole-body interactions, and accuracy to targets. Motion tracking-driven reinforcement learning (RL) provides stability in whole-body movement coordination, but a fixed reference makes it hard to adapt to varied ball positions and strike timings; in contrast, task reward-driven RL struggles to explore and discover valid kicks from scratch. We therefore introduce RoboNaldo, a three-stage motion-guided curriculum RL framework for high-impulse humanoid interaction. A single human-kick reference is used as a scaffold and progressively shifts optimization towards shooting performance. The curriculum first learns a stable whole-body kicking prior, then adapts the kick to free-kick settings where the ball is stationary at random positions, and finally extends it to moving-ball shooting through a locomotion-command and kick-trigger interface. A high-level heuristic planner controls this interface during training, while alternative high-level controllers can drive the same low-level policy at inference. In simulation, RoboNaldo demonstrates free-kick shot error 48.6% lower and shoot velocity 2.96x than prior work baselines. In real world on a Unitree G1 with onboard perception, RoboNaldo attains 0.73 m and 0.86 m average target shooting error from 3 m away in free-kick and moving-ball cases, accordingly. And the post-contact ball velocity reaches 13.10 m/s, which is 59-71% of reported professional open-play shot speed. Project page: https://opendrivelab.com/RoboNaldo.
Gravitational waves and ultralight dark matter are among the most compelling frontiers in fundamental physics, motivating proposals for very-long-baseline atom interferometerssuch as AION1, MAGIS2, AICE3 and AEDGE4 that aim to detect at frequencies at which ground-based5 and space-borne6 laser interferometers lose sensitivity. Very-long-baseline atom interferometers look for signals by comparing the quantum phase evolution of widely separated atomic ensembles interrogated by a common laser. However, their performance depends critically on suppressing noise sources, particularly laser phase noise. The experimental validation of such noise rejection remains an important challenge. Here we demonstrate a prototype differential atom interferometer based on the single-photon clock transition of fermionic 87Sr. Thus, we obtain a gradiometer configuration with a species intrinsically suited to kilometre-scale and space-baseline operation. The instrument operates at the standard quantum limit7 with no excess noise beyond atom shot noise. The differential configuration maintains quantum-limited sensitivity in the presence of several radians of artificially injected laser phase noise per shot, which emulates the conditions expected in a very-long-baseline atom interferometer. We also demonstrate the recovery of coherent oscillatory signals across a broad frequency range under fully phase-randomized conditions, a capability that is inaccessible to a single interferometer operating in the same regime. These results provide an experimental validation of the noise-immune measurement principle underlying very-long-baseline atom interferometers and mark an important step towards next-generation quantum sensors for gravitational-wave detection and searches for ultralight dark matter8,9. A prototype differential atom interferometer operates at the standard quantum limit with no excess noise beyond atom shot noise, achieving performance in line with the specifications for future long-baseline atom interferometers.
Background: Recruitment of underrepresented populations, including Black and Hispanic populations, for Food is Medicine (FIM) and cardiovascular trials, may pose significant challenges. Methods: We implemented a multi-component recruitment approach for the THRIVE (AdapTive personalized dietitian coacHing and messaging with pRoduce prescrIptions to improVE healthy dietary behaviors) pilot trial to engage primarily Black and Hispanic adults in a Food is Medicine for hypertension intervention. The recruitment approaches included community engagement at approximately 40 community events (cultural festivals and neighborhood gatherings); partnerships with 8 community and faith-based service hubs and food distribution sites; recruitment through safety net primary care clinics, digital outreach via the study website, and social media campaigns; and direct recruitment at places of worship. We report lessons learned from the community engagement process, recruitment efficiency, representativeness, and retention outcomes. Results: Within 6 months, the enrollment target was exceeded by 40%, with an accrual index of 1.04. Over 1,000 individuals were reached through the direct-to-community engagement process, while faith-based partnerships engaged about 900 adults. There were 2,673 visits to the study webpage, and social media achieved 12,259 impressions with 399 clicks. About 95% of participants resided within 10 miles of the faith-based recruitment sites. Face-to-face engagement at the food distribution sites within faith-based organizations or community service hubs outperformed digital methods. Faith leader endorsements and follow-up in-person meetings (following unsuccessful email outreach) dramatically increased recruitment. Regarding retention, pre-randomization attrition was 6%, and 82% of participants completed the study. Conclusion: Culturally tailored, community-engaged recruitment grounded in faith-based and local community partnerships, was highly effective in engaging Black and Hispanic populations in this FIM cardiovascular trial. This provides a replicable model for implementing equitable and sustainable cardiovascular health interventions.
Mixed reality applications can be designed for hand rehabilitation. Augmented reality (AR) head mounted displays (HMDs) specifically allow for ecologically valid tasks because individuals can see their real environment and interact with real objects while receiving additional cues on the HMD. While these applications rely on accurate hand pose estimation, there is a gap in investigating the influence of hand impairment or occlusion from real-object interactions on pose estimation accuracy. Further, comparisons between AR HMD predictions and state-of-the-art pose estimation methods have not been established. The current study assessed pose estimation accuracy of the HoloLens 2 HMD and state-of-the-art pose estimation algorithms (WiLoR, HaMeR, WildHands, and MediaPipe) while individuals with cervical spinal cord injury (cSCI; n = 13, Neurological Level of Injury: C3-C6; American Spinal Injury Association Impairment Scale: A-D) and 15 uninjured controls interacted with clear and opaque objects. Ground truth estimates of 3D joint positions were generated via triangulation from a multi-camera setup. Pose estimation accuracy did not differ between the cSCI and uninjured control groups suggesting that 3D joint predictions from the HoloLens 2 and pose estimation algorithms can generalize to populations with hand impairment. Further, clear objects provided a small accuracy advantage over opaque objects (0.1 mm) and predictions from both WiLoR and HaMeR were slightly more accurate than the HoloLens 2 (2 mm). Overall, these results suggest that the HoloLens 2 may be viable for hand rehabilitation applications and the dataset generated can be used to refine pose estimation methods for hand-impaired populations.
arXiv:2606.12563v1 Announce Type: new Abstract: Arbor is a multi-agent framework that introduces structured tree search as a cognition layer for autonomous agents operating in large, stateful action spaces. Prior autonomous optimization systems operate on isolated targets with stateless evaluation. Arbor instead maintains an explicit search tree of scored hypotheses that serves as the shared working memory across agents, evolving with every measurement, treating failures as diagnostic signal that reshapes subsequent exploration, and expanding as prior successes shift the bottleneck distribution. We validate Arbor on full-stack LLM inference optimization, a domain where achieving peak performance has historically required coordinated effort from engineering teams across the application, framework, compiler, kernel, and hardware stack. Arbor pairs an Orchestrator agent, which drives optimization by delegating to Domain Specialists across the inference stack, with a Critic agent that safeguards stability through root-cause analysis, introspection, and measurement validation – a checks-and-balances architecture where neither agent can unilaterally drive the system. Agent capabilities are decomposed into hard skills (domain expertise) and soft skills (coordination protocols that determine how contributions compose), enabling fully autonomous multi-day campaigns. Arbor achieves up to 193% inference throughput-latency Pareto improvement over vendor-optimized baselines, while a single agent without the harness plateaus at +33% throughput improvement and crashes irrecoverably within hours. Arbor generalizes to multiple generations of hardware platform, and run-to-run variance is within 2 percentage points demonstrating that the method is hardware-agnostic and reproducible.
We present WireframeDETR, our submission to the Structured Semantic 3D Reconstruction (S23DR) 2026 Challenge, which requires predicting a 3D building wireframe from multi-view COLMAP point clouds. Our method applies DETR-style set prediction directly to 3D point clouds, producing wireframes as sets of edge coordinate pairs without any intermediate vertex detection stage. We introduce three technical contributions: (1) contrastive denoising training that stabilises noisy Hungarian matching in early epochs; (2) a multi-scale encoder that aggregates the last encoder layer outputs via learned scalar weights; and (3) progressive auxiliary loss weighting that concentrates gradient signal on the decoder layers that most benefit from it. Our model achieves a public test HSS of 0.575 (F1~=~0.664, IoU~=~0.516) and a best validation HSS of 0.534 on the cleaned val split.
arXiv:2605.12513v2 Announce Type: replace-cross Abstract: Influence maximization (IM) in real platforms is challenged by incomplete, noisy social graphs and non-stationary diffusion dynamics. We propose SP-GCRL, a social-propagation-aware graph contrastive reinforcement learning framework that learns end-to-end seed selection under partial observability.We first introduce a social-propagation-aware nonlinear diffusion function to model reinforcement/diminishing effects and probability drift under repeated exposure; we then construct dual structural views and perform contrastive learning to obtain node representations robust to missing edges and weak ties, while replacing expensive strategy metrics with a GAT-based regression surrogate to improve efficiency and scalability; finally, we use DDQN to learn an end-to-end seed selection policy on top of these representations. Experiments on multiple real-world networks show that SP-GCRL achieves significant gains over heuristic and learning-based baselines across budgets and topologies, while maintaining strong large-scale scalability.
arXiv:2606.01561v2 Announce Type: replace Abstract: Aligning Large Language Models (LLMs) with human preferences is often formulated via Direct Preference Optimization (DPO). However, the standard Bradley-Terry instantiation of DPO is limited in modeling common departures from transitivity in human preferences. To address this, recent work has introduced Self-Play Preference Optimization (SPPO), which iteratively refines the policy by training on self-generated win-lose pairs. Our investigation, however, reveals a critical instability in SPPO: the optimization is prone to policy degeneration when the preference oracle assigns overly confident wins to semantically indistinguishable responses. To mitigate this, we propose S-SPPO, a dual-space semantic calibration framework comprising: i) Supervision Calibration via semantic gating, which anneals win rate targets toward the maximum-entropy baseline as semantic overlap increases; and ii) Representation Calibration via latent repulsion to enforce geometric diversity to prevent manifold collapse and maintain latent diversity between chosen and rejected samples. Theoretically, we show that the calibration preserves the constant-sum game structure, facilitating convergence to a Nash Equilibrium. Empirically, S-SPPO avoids the performance degradation seen in prior methods, achieving 52.19% win rate and 47.46% length-controlled win rate on AlpacaEval 2.0 with Llama-3-8B, without using additional human-annotated preferences during training. The code will be available at https://github.com/xiwenc1/s-sppo.
In multimodal large language models (MLLMs), inference cost is largely dominated by the visual token prefix rather than the language backbone, making token reduction a key factor for improving efficiency. Existing approaches typically assign independent importance scores to visual tokens and retain a fixed number of top-ranked tokens, implicitly assuming token independence and a uniform compression ratio across inputs. In this work, we reformulate visual token pruning as a sequential decision-making process. Specifically, we introduce a pointer-style selection mechanism that iteratively chooses informative tokens, conditioning each decision on previously selected ones, and dynamically determines when to stop via a learned termination action. This enables joint optimization of both the selected subset and its size. To enable end-to-end training under standard language modeling objectives, we design a differentiable relaxation based on a variance-preserving noise interpolation scheme, allowing gradients to propagate through the discrete selection process. Extensive experiments on LLaVA-v1.5-7B and Qwen2.5-VL-7B demonstrate that our approach consistently outperforms fixed-ratio baselines across different compression levels. Under aggressive pruning that removes 88.9% of visual tokens, our method preserves 94.6% of the original accuracy while achieving a 1.88x speed-up in prefill latency.
arXiv:2606.15930v1 Announce Type: cross Abstract: Simulation is central to validating autonomous driving systems, yet current pipelines are limited by insufficient scenario diversity due to costly High Definition (HD) map creation. Scaling HD maps requires expensive data collection and manual processing. Moreover, existing generative models lack the fine-grained control necessary to target specific road topologies during generation. This paper presents a data-driven pipeline for controllable HD map generation using latent diffusion and ControlNet for spatial conditioning. To our knowledge, we are the first to inject spatial guidance signals into a diffusion model for HD map synthesis. Furthermore, our model supports adjustable conditioning strength through classifier-free guidance and city-level style transfer via city label conditioning. To complement existing metrics, we introduce two novel metrics to evaluate adherence to the control signal and similarity to ground-truth maps. Experiments demonstrate that our model generates realistic HD maps that faithfully follow input road topologies while accurately preserving city-specific details.
GPU kernel optimization represents a paradigm where functional correctness is assumed and execution efficiency is the objective. We present daVinci-kernel, a reinforcement learning framework that couples skill discovery with skill exploitation through a dynamically evolving skill library. daVinci-kernel jointly trains three agents sharing one LLM backbone: a Skill Selection Agent that retrieves relevant techniques via BM25 and LLM reranking, a Policy Agent that generates multi-turn CUDA/Triton kernels conditioned on selected skills, and a Skill Summary Agent that distills successful rollouts into reusable skills. Candidate skills are added only after execution-based verification confirms reproducible speedups. All three agents share a single LLM backbone, are initialized via a structured SFT cold start on diversity-filtered data, and are then jointly optimized end-to-end with multi-turn REINFORCE and per-agent advantage estimation. On KernelBench, daVinci-kernel-14B achieves 37.2%, 70.6%, and 32.2% on Level 1, Level 2, and Level 3 under the Fast$_1$ threshold, outperforming the strongest prior RL-trained model, Dr.Kernel-14B.
Non-invasive brain-computer interfaces suffer severe fidelity degradation in neural visual decoding when generalizing to natural visual experiences. Conventional multimodal contrastive representation learning solely optimizes geometric distance alignment, neglecting semantic consistency and subject selectivity, causing spurious zero-shot alignment. We propose SUP-MCRL, a unified framework integrating three collaborative mechanisms: (1) Semantic-entity Aware Visual Encoder (SAVE), learning spatial attention to extract semantic content without pre-trained saliency models; (2 Unified EEG Enhancer (UEE), employing multi-scale atrous convolutions and inter-band attention for adaptive cross-subject robustness; and (3) Prototype-based Progressive Augmenter (PPA), maintaining an EMA-updated pseudo-feature pool to prevent representation collapse. Zero-shot experiments on THINGS-EEG achieve 66.0%/91.9% (Top-1/Top-5) intra-subject and 24.0%/52.9% LOSO accuracy, surpassing state-of-the-art methods. Code is available at https://github.com/NZWANG/SUP-MCRL.
We introduce SMART, a framework for learning a flexible, interpretable, and scalable spatio-temporal brain atlas from longitudinal high-resolution 3D medical images. Existing approaches to spatio-temporal atlas construction rely on black-box generative models that lack flexibility, limit interpretability, and struggle to scale to high-dimensional data. SMART addresses these challenges by learning a continuous disease-time atlas that decouples global group-wise disease dynamics from their patient-specific anatomical manifestation. Guided by anatomically inspired priors, SMART models interpretable global trajectories of regional progression along a shared disease timeline through region-specific differential equations. Global trajectories are further personalized to individual anatomies via dense diffeomorphic displacements parameterized by a flexible and scalable multi-scale Neural Cellular Automata. Evaluated on five longitudinal MRI datasets in Alzheimer's disease (ADNI-1/GO/2, OASIS-3, AIBL; > 1,300 subjects), SMART produces anatomically meaningful predictions of disease progression and achieves state-of-the-art forecasting accuracy and improved temporal consistency over adversarial and diffusion baselines. Our approach establishes a new paradigm for flexible, interpretable, and scalable modeling of spatio-temporal change in high-dimensional medical image time-series.
arXiv:2606.11918v1 Announce Type: new Abstract: Current Large Reasoning Models (LRMs) exhibit remarkable general capabilities but significantly underperform in spatial reasoning tasks. Existing approaches treat this gap as a knowledge deficit, relying on supervised fine-tuning (SFT) to ingest labeled spatial data from external vision sources or synthetic engines. In contrast, we argue that for many tasks, spatial reasoning capabilities are already present in pre-trained LRMs but require alignment through logical coherence under geometric 2D and 3D constraints. In this work, we propose a self-supervised reinforcement learning (RL) framework that targets the internal reasoning process without requiring ground-truth annotations. By formalizing the notion of consistency verifiers – reward functions that check for geometric and semantic consistency under transformations – we demonstrate that models can improve their spatial reasoning abilities. We use both image transformations, like flipping, and textual transformations, like swapping the order of objects in the question, and propose a new optimal transport-based RL strategy, OT-GRPO, which is a minimal-matching variant of group relative policy optimization tailored to pairwise verifiers. We show that this label-free consistency training approaches the accuracy of models trained with ground-truth supervision and achieves similar generalization across diverse tasks and data domains.
Large language models (LLMs) are increasingly used for clinical text tasks such as summarization and revision. While most studies evaluate the fluency and coherence of LLM-generated text, whether LLMs correctly preserve diagnostic uncertainty remains underexplored. In clinical practice, phrases such as ``possible pneumonia'' communicate the strength of available evidence and directly guide decisions about follow-up testing and treatment. Altering these uncertainty expressions can change the clinical meaning entirely. In this paper, we systematically evaluated this problem in two steps. First, we constructed a benchmark of 1,200 clinical documents with 9,184 uncertainty annotations across five levels. Second, we evaluated three LLMs on this benchmark. Our results show that (1) LLMs preserve the original uncertainty cues poorly, often less than half the time; (2) LLMs struggle with nuanced distinctions between adjacent levels. This work reveals a failure mode not captured by standard evaluation metrics and provides implications for the safe deployment of LLMs in clinical workflows.
Multimodal Large Language Models (MLLMs) remain unreliable on fine-grained visual perception, even when high-resolution inputs preserve the necessary local details. We identify this limitation as visual context rot: decisive evidence may exist in the full image, yet fail to be reliably selected and used amid redundant visual context. We propose LOCUS (LOcal visual CUe Search), a training framework that teaches MLLMs to internalize local evidence search through a verifiable proxy task. During training, LOCUS provides a local crop as a visual cue and optimizes the model to recover its spatial support in the full image using an IoU-based reward. The visual cue is used only during training, leaving the standard image-question inference interface unchanged. Experiments across fine-grained perception, hallucination, general understanding, and reasoning benchmarks show that LOCUS improves localization-sensitive visual understanding while preserving broad capabilities. Attention analyses further indicate stronger focus on task-relevant evidence regions, suggesting that training-time visual cue search provides an effective route to internalized fine-grained evidence selection.
Objectives: Amblyopia is a pediatric visual disorder traditionally treated by patching the fellow eye, though many patients retain residual amblyopia post-treatment. Increasing evidence suggests that visual plasticity allows treat-ment beyond the classical therapeutic window. AMBER evaluated the efficacy of binocular serious games in virtual reality (VR) in residual amblyopia. Methods and Analysis: The monocentric, prospective, randomized, crossover trial (reported as case series) includ-ed 14 anisometropic, strabismic, or mixed residual amblyopia patients (6-35 years; 5 children, 9 adults). Participants underwent two 2-month intervention phases: optical correction (standard care) and standard care plus VR games (2.5 h/week), each with a 2-month follow-up. Best-corrected visual acuity (BCVA), stereoacuity, and reading speed were assessed (5 timepoints) using the Sloan and Landolt charts, the Titmus, TNO, Lang II, Asteroid, and Mnread tests. Compliance and adverse events (AE) were recorded. Results: VR training improved BCVA in 10 amblyopic eyes (Landolt and Sloan), with more pronounced effects in anisometropic patients. Six patients showed improved stereoacuity (Titmus; 4x mixed, 1x anisometropic, 1x stra-bismic amblyopia), persistent only in children (1x strabismic, 1x mixed amblyopia). Four improvements were ob-served with TNO (1x), Lang II (1x), Asteroid (0x), and MNread (1x). Despite positive trends, when comparing re-sults of individual patients, between both eyes, and with standard treatment, consistency of improvements cannot be conclusively demonstrated. One non-severe AE (dizziness) was reported. Conclusions: Following individual cases, VR training improved BCVA and stereoacuity, particularly in children and patients with high compliance. However, considering the cohort as a whole, consistency of effects has to be confirmed in larger groups. Thus, the methodologically sophisticated AMBER study revealed differences in VR treatment efficacy between amblyopia types, children/adults, endpoints and tests, offering precious data for the design of meaningful future studies. It shows that neurovisual plasticity gauged by VR-games offers safe, engaging treatment options for residual amblyopia.