Standardized pre-consultation by a large language model agent vs ophthalmology residents: a randomized clinical trial

By: Ming-Jie Luo · Yunxi Lai · Wei Wang · Shaowei Bi · Ling Jin · Jingjing Chen · Xiaohang Wu · Zhenzhe Lin · Jingrui Wang · Jianyu Pang · Meimei Dongye · Yue Wu · Zhenzhen Liu · Haotian Lin

5 October 2026 at 08:00

npj Digital Medicine, Published online: 05 October 2026; doi:10.1038/s41746-026-03232-x

Urine cell-free RNA for bladder cancer detection and treatment response prediction

MRD

By: Kevin J Liu · William Y Shi · Monica C Nesselbush · Nick A Phillips · Kathleen E Mach · Bogdan Luca · Ravi K Bajpai · Isabel Jabara · Jordan C Cheng · Takeshi Sugio · Navika Shukla · Emily C Shen · Ivy Lee Koh · Douglas J Wu · Chih-Long Liu · Diego Almanza · Soyeong Jun · Feng Tian · Emily G Hamilton · Christian R Hoerner · Rui Wang · Mohammad S Esfahani · Angela Hui · Vinh La · Grace Holton · Gabriela Rodriguez · Timothy J Lee · Xi Ling · Nikitha Kalahasti · Spoorthi Vallamkonda · Patrick Mullane · Simon B Chen · Eugene Shkolyar · Kris Prado · Ali Raza Khaki · Emily Chan · Mark K Buyyounouski · Alice C Fan · Eila C Skinner · Ash A Alizadeh · Joseph C Liao · Maximilian Diehn

2 October 2026 at 18:00

Nat Med. 2026 Oct 2. doi: 10.1038/s41591-026-04673-3. Online ahead of print.

ABSTRACT

Urine biomarkers promise to improve noninvasive detection and molecular characterization of genitourinary malignancies. Here we describe urine random priming and affinity capture of cell-free RNA (cfRNA) fragments for enrichment analysis by sequencing (uRARE-seq), a liquid biopsy method for urine cfRNA profiling, and apply it to 683 urine samples from patients with cancer and controls. Urine cfRNA contained transcripts from genitourinary tissues and, in patients with prostate, kidney or bladder cancer, tumor-derived transcripts. uRARE-seq demonstrated 95% sensitivity at 90% specificity for detecting localized bladder cancer. The method outperformed urine tumor DNA analysis and was unaffected by the presence of field-effect mutations. Urine cfRNA analysis also sensitively detected minimal residual disease and distinguished complete molecular responses after surgery from those after intravesical Bacillus Calmette-Guérin (BCG). Pretreatment urine from complete responders to BCG was enriched for T cell and other immune signatures, suggesting a preexisting antitumor immune response, whereas nonresponders showed higher expression of proliferation-related genes. In pretreatment urine from 114 patients, this biological difference enabled development of a biomarker predicting likelihood of response to BCG versus chemotherapy (area under the curve 0.93) that was strongly associated with risk of recurrence. Urine cfRNA analysis is therefore a promising biomarker approach for bladder cancer and potentially other urologic malignancies, although prospective studies are needed to assess its clinical utility.

PMID:42827132 | DOI:10.1038/s41591-026-04673-3

Rational and computation-assisted engineering of a compact and efficient CRISPR–Cas12f genome editor

Molecular Therapy

By: Jingtong Liu · Shengzhou Wang · Bei Wang · Tonglu Yu · Xuan Liu · Miaomiao Li · Min-Chi Ku · Rui Wang · Renjie Chai · Wei V. Zheng · Yongming Wang

31 August 2026 at 08:00

Structure-guided design combined with protein language model-guided engineering and sgRNA optimization enables the development of a compact and highly efficient CRISPR–Cas12f genome editor. This integrated strategy substantially improves genome-editing activity while preserving high specificity, expanding the therapeutic potential of compact CRISPR systems.

CityPlanner: A Sandbox Agent for Executable Urban Planning

cs.AI, q-bio.NC updates on arXiv.org

By: Wentao Zhang · Jingyuan Wang · Zetong Zhou · Yifan Yang · Wenrui Wang

10 September 2026 at 12:00

arXiv:2609.09578v1 Announce Type: new Abstract: Urban planning is a real-world spatial optimization problem that requires selecting feasible actions from large candidate spaces under practical objectives such as cost and service quality. Existing optimization and reinforcement learning methods are effective for fixed formulations, but often depend on task-specific representations and constraint handling. We propose \emph{CityPlanner}, a sandbox-agent framework for executable urban planning. CityPlanner introduces \emph{UrbanSandbox}, a unified file-based environment where agents inspect task files, generate plans, run evaluators, and revise decisions based on executable feedback. To make learning tractable, we further propose atomic-task reinforcement learning, which decomposes long sandbox trajectories into \emph{BuildPlan} for initial construction and \emph{ImprovePlan} for feedback-based refinement. Experiments on a real-world benchmark show that CityPlanner consistently outperforms heuristic, task-specific RL, and general LLM-agent baselines. Ablations verify the contributions of UrbanSandbox, atomic-task RL, and iterative deployment. We release the code and dataset at https://anonymous.4open.science/r/co-agent-C1C8

Dear Algo: A Precision-First Agentic Intent Layer for Unified Search and Recommendation

cs.AI, q-bio.NC updates on arXiv.org

By: Rui Wang · Jiazhou Wang · Zheng Wei · Chenglin Lu · Fangcheng Sun · Ivy Sun · Jin Sun · Hui Geng · Lillian Zhang · Chao Yang · Lei Chen · Shahin Sefati · Reem Helou · Joe Zhou · Babak Shakibi · Yiyi Pan · Bi Xue · Hong Yan · Shujian Bu

10 September 2026 at 12:00

arXiv:2608.15877v3 Announce Type: replace Abstract: Search and recommendation serve a shared discovery objective but encode intent differently. We study this boundary through Dear Algo on Threads, a deployed product where open-ended requests such as \emph{more NBA news} or \emph{less politics} steer subsequent feed recommendations rather than return a one-shot result list. Its agentic intent layer compiles explicit, inferred, negative, and compound intent into a grounded executable plan, then invokes conventional retrieval and optional semantic or multimodal reranking. The layer shares an intent-to-retrieval contract without requiring one model or serving path across search-like and recommendation-like modes. We evaluate Dear Algo under a precision-first objective. In a blinded audit of 300 public request-item pairs (296 evaluable), a strict categorical LLM-as-a-judge gate achieved 94.4\% exact-Relevant precision [88.8\%, 98.9\%]. Across 72 normalized request clusters, the full configuration produced 7.73 judge-qualified candidates per 20 slots versus 6.61 for an LLM-derived-query baseline, a gain of 1.11 [0.12, 2.12]. In a candidate-randomized serving-path study restricted to the reranker path's first 72 eligible hours, the user-weighted judge-Irrelevant share among judged admissions was 2.80\% versus 4.78\% off (-1.97 points [-3.02, -0.94]), while Exact-Relevant share was 2.24 points higher [0.08, 4.41]. Together, these studies show how explicit natural-language intent can be carried into feed recommendation under a precision-first evaluation framework

AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting

cs.AI, q-bio.NC updates on arXiv.org

By: Rui Wang · Renhao Xue · Ray Razi · Huan Song · Hannah R. Marlowe

26 May 2026 at 12:00

arXiv:2605.25166v1 Announce Type: cross Abstract: Time series forecasting models are increasingly scaled through large Transformer backbones, yet most existing approaches process all series through a shared dense computation path despite substantial heterogeneity in temporal structure. Mixture-of-Experts (MoE) offers a natural alternative by enabling conditional computation, but standard MoE routing leaves expert specialization weakly identified and often unstable during downstream adaptation. We propose AME-TS, a structure-guided sparse time series foundation model that aligns expert routing with interpretable temporal structure. AME-TS first uses a lightweight regime predictor to estimate series-level descriptors, including forecastability, seasonality, trend, and sparsity, and maps them to a soft structural prior over experts. This series-level prior guides token-level routing during training, encouraging structure-aligned specialization. On the GIFT-Eval benchmark, AME-TS delivers a strong accuracy-efficiency tradeoff across model scales: it substantially outperforms existing time series foundation models at small model scales and remains competitive with the strongest models at larger scales, while activating substantially fewer parameters through sparse routing. We further show that AME-TS learns more interpretable routing geometry and substantially more stable expert specialization than standard MoE during fine-tuning on the M5 dataset. These results suggest that structure-aware routing is an effective and reliable way to realize the benefits of sparse expert models for time series forecasting.

Multi-omics integration and Mendelian randomization reveal the mechanisms and experimental validation of curcumin targeting the RXRA-PI3K/AKT axis to enhance cisplatin sensitivity in gastric cancer

Omics in Gastric

By: Xiaoran Sun · Na Wu · Xue Bai · Xiang Zhang · Rui Wang · Shuai Du · Li Liu · Duo Li

1 May 2026 at 18:00

Front Oncol. 2026 Apr 15;16:1791971. doi: 10.3389/fonc.2026.1791971. eCollection 2026.

ABSTRACT

OBJECTIVE: This study aimed to integrate multi-omics analyses with genetic causal inference to identify key genes associated with cisplatin resistance in gastric cancer and to evaluate the potential mechanism by which curcumin enhances cisplatin sensitivity through relevant pathways.

METHODS: Cisplatin resistance-related transcriptomic datasets(GSE14210 and GSE31811) and a gastric cancer single-cell transcriptomic dataset (GSE183904) were obtained from the Gene Expression Omnibus(GEO)database. Differential expression analysis was performed to identify resistance-associated differentially expressed genes(DEGs),followed by GO and KEGG enrichment analyses. Putative curcumin targets were collected and intersected with DEGs to obtain candidate genes. Mendelian randomization (MR) analysis was conducted using the TwoSampleMR framework to evaluate the genetic association between RXRA expression and gastric cancer risk, with robustness and sensitivity analyses based on multiple MR methods. RXRA expression was further evaluated, along with pathway activity assessment using GSEA and GSVA, and molecular docking was performed to explore the potential binding of curcumin to RXRA. In vitro experiments were performed using the cisplatin-resistant gastric cancer cell lineNCI-N87/DDP. Drug effects and chemosensitization under combination treatment were assessed by CCK-8 assays, synergy was evaluated using the combination index(CI),and changes in key proteins in thePI3K/AKT pathway were measured by Western blotting.

RESULTS: A total of 595 DEGs associated with cisplatin resistance were identified. Functional enrichment analyses indicated that these DEGs were mainly involved in extracellular matrix remodeling and adhesion, secretion and vesicular transport, and signaling pathways including PI3K-Akt.The intersection of curcumin targets with DEGs highlighted RXRA as a key candidate gene. MR results indicated that genetically predicted increased RXRA expression was significantly associated with elevated gastric cancer risk (OR = 4.216,95%CI:1.201-14.797,P=0.025). GSEA and GSVA suggested that high RXRA expression was associated with altered activity of pathways related to lysosome, proteasome, oxidative phosphorylation, and the pentose phosphate pathway. Single-cell analysis indicated that RXRA was mainly expressed in tissue stem cells and fibroblasts. Molecular docking predicted a feasible interaction between curcumin and RXRA. In vitro experiments demonstrated that curcumin inhibited the viability of resistant cells and showed a synergistic trend when combined with cisplatin. Western blotting revealed decreased p-PI3K and p-AKT levels following curcumin treatment, supporting an inhibitory effect on the PI3K/AKT pathway.

CONCLUSION: These findings highlight RXRA as a candidate gene associated with cisplatin resistance-related programs in gastric cancer. Curcumin may enhance cisplatin sensitivity by influencing RXRA-associated transcriptional networks and suppressing PI3K/AKT signaling. This study provides new candidate targets and experimental evidence for mechanistic investigation and combination treatment strategies to overcome cisplatin resistance in gastric cancer.

PMID:42063729 | PMC:PMC13124633 | DOI:10.3389/fonc.2026.1791971

Correction: Steroid receptor coactivator-1 facilitates METTL3-mediated m6A modification by coactivating NF-κB and promotes the malignant progression of glioblastoma

Oncogene - Issue - nature.com science feeds

By: Liang Liu · Rui Wang · Ke Cheng · Chunmei Bai · Yuke Ji · Yifei Zhang · Haoran Yang · Miaomiao Gong · Fang Xie · Yongshun Zhao · Jinjin Pan · Yuhui Yuan

15 April 2026 at 08:00

Oncogene, Published online: 15 April 2026; doi:10.1038/s41388-026-03788-8

Correction: Steroid receptor coactivator-1 facilitates METTL3-mediated m6A modification by coactivating NF-κB and promotes the malignant progression of glioblastoma

Transplantation of encapsulated mitochondria alleviates dysfunction in mitochondrial and Parkinson’s disease models

Cell

By: Shiwei Du · Qi Long · Yanshuang Zhou · Jiangqin Fu · Hao Wu · Liang Yang · Yaohang Xie · Yingzhe Ding · Maolei Zhang · Jingyi Guo · Mengfei Wang · Jiajun Lin · Mingli Hu · Jian Zhang · Deyang Yao · Wei Li · Feixiang Bao · Ge Xiang · Yi Wu · Yile Huang · Haozhao Liang · Rui Wang · Heying Li · Baodan Chen · Chong Li · Junwei Wang · Jiwei Zhang · Dajiang Qin · Jianwei Sun · Yun Zhu · Fei Sun · Wuming Wang · Gang Lu · Wai-Yee Chan · Hui Zhao · Chenli Liu · Xingguo Liu

18 March 2026 at 08:00

A mitochondrial transplantation approach rescues mitochondrial deficiency and prevents mitochondrial DNA depletion syndrome, Leigh syndrome, and Parkinson’s disease in cellular and mouse models.

XModBench: Benchmarking Cross-Modal Capabilities and Consistency in Omni-Language Models

cs.AI, q-bio.NC updates on arXiv.org

By: Xingrui Wang · Jiang Liu · Chao Huang · Xiaodong Yu · Ze Wang · Ximeng Sun · Jialian Wu · Alan Yuille · Emad Barsoum · Zicheng Liu

7 April 2026 at 12:00

arXiv:2510.15148v2 Announce Type: replace-cross Abstract: Omni-modal large language models (OLLMs) aim to unify audio, vision, and text understanding within a single framework. While existing benchmarks primarily evaluate general cross-modal question-answering ability, it remains unclear whether OLLMs achieve modality-invariant reasoning or exhibit modality-specific biases. We introduce XModBench, a large-scale tri-modal benchmark explicitly designed to measure cross-modal consistency. XModBench comprises 60,828 multiple-choice questions spanning five task families and systematically covers all six modality compositions in question-answer pairs, enabling fine-grained diagnosis of an OLLM's modality-invariant reasoning, modality disparity, and directional imbalance. Experiments show that even the strongest model, Gemini 2.5 Pro, (i) struggles with spatial and temporal reasoning, achieving less than 60% accuracy, (ii) reveals persistent modality disparities, with performance dropping substantially when the same semantic content is conveyed through audio rather than text, and (iii) shows systematic directional imbalance, exhibiting lower consistency when vision serves as context compared to text. These findings indicate that current OLLMs remain far from truly modality-invariant reasoning and position XModBench as a fundamental diagnostic tool for evaluating and improving cross-modal competence. All data and evaluation tools will be available at https://xingruiwang.github.io/projects/XModBench/.

MindCube: Spatial Mental Modeling from Limited Views

cs.AI, q-bio.NC updates on arXiv.org

By: Qineng Wang · Baiqiao Yin · Pingyue Zhang · Jianshu Zhang · Kangrui Wang · Zihan Wang · Jieyu Zhang · Keshigeyan Chandrasegaran · Han Liu · Ranjay Krishna · Saining Xie · Jiajun Wu · Li Fei-Fei · Manling Li

1 April 2026 at 12:00

arXiv:2506.21458v2 Announce Type: replace Abstract: Can Vision-Language Models (VLMs) imagine the full scene from just a few views, like humans do? Humans form spatial mental models naturally, internal representations of unseen space, to reason about layout, perspective, and motion. Our MindCube benchmark with 21,154 questions across 3,268 images exposes this critical gap, where existing VLMs exhibit near-random performance. Using MindCube, we systematically evaluate how well VLMs build robust spatial mental models through representing positions (cognitive mapping), orientations (perspective-taking), and dynamics (mental simulation for "what-if" movements). We then explore three approaches to help approximate spatial mental models in VLMs, focusing on incorporating unseen intermediate views, natural language reasoning chains, and cognitive maps. The significant improvement comes from a synergistic approach, "map-then-reason", that jointly trains the model to first generate a cognitive map and then reason upon it. By training models to reason over these internal maps, we boosted accuracy from 37.8% to 57.8% (+20.0%). Adding reinforcement learning pushed performance even further to 61.3% (+23.5%). Our key insight is that such scaffolding of spatial mental models, actively constructing and utilizing internal structured spatial representations with flexible reasoning processes, significantly improves understanding of unobservable space.

When Metrics Disagree: Automatic Similarity vs. LLM-as-a-Judge for Clinical Dialogue Evaluation

cs.AI, q-bio.NC updates on arXiv.org

By: Bian Sun · Zhenjian Wang · Orvill de la Torre · Zirui Wang

1 April 2026 at 12:00

arXiv:2603.00314v2 Announce Type: replace-cross Abstract: As Large Language Models (LLMs) are increasingly integrated into healthcare to address complex inquiries, ensuring their reliability remains a critical challenge. Recent studies have highlighted that generic LLMs often struggle in clinical contexts, occasionally producing misleading guidance. To mitigate these risks, this research focuses on the domain-specific adaptation of \textbf{Llama-2-7B} using the \textbf{Low-Rank Adaptation (LoRA)} technique. By injecting trainable low-rank matrices into the Transformer layers, we efficiently adapted the model using authentic patient-physician transcripts while preserving the foundational knowledge of the base model. Our objective was to enhance precision and contextual relevance in responding to medical queries by capturing the specialized nuances of clinical discourse. Due to the resource-intensive nature of large-scale human validation, the model's performance was evaluated through a dual-track framework: \textbf{Track A} utilized traditional lexical similarity metrics (e.g., BLEU, ROUGE), while \textbf{Track B} employed an "LLM-as-a-Judge" paradigm using GPT-4 for semantic assessment. Our results demonstrate that while the LoRA-enhanced model achieved significant improvements across all quantitative lexical dimensions, a profound disagreement surfaced in the GPT-4 evaluation, which marginally favored the baseline model's conversational flow. This metric divergence underscores a pivotal finding: traditional automated scores may not fully reflect clinical utility. Consequently, we propose that while automated metrics and LLM judges serve as valuable developmental proxies, rigorous validation by human medical experts remains an indispensable requirement for the safe deployment of LLMs in healthcare settings.

PersonalQ: Select, Quantize, and Serve Personalized Diffusion Models for Efficient Inference

cs.AI, q-bio.NC updates on arXiv.org

By: Qirui Wang · Qi Guo · Yiding Sun · Junkai Yang · Dongxu Zhang · Shanmin Pang · Qing Guo

25 March 2026 at 12:00

arXiv:2603.22943v1 Announce Type: new Abstract: Personalized text-to-image generation lets users fine-tune diffusion models into repositories of concept-specific checkpoints, but serving these repositories efficiently is difficult for two reasons: natural-language requests are often ambiguous and can be misrouted to visually similar checkpoints, and standard post-training quantization can distort the fragile representations that encode personalized concepts. We present PersonalQ, a unified framework that connects checkpoint selection and quantization through a shared signal -- the checkpoint's trigger token. Check-in performs intent-aligned selection by combining intent-aware hybrid retrieval with LLM-based reranking over checkpoint context and asks a brief clarification question only when multiple intents remain plausible; it then rewrites the prompt by inserting the selected checkpoint's canonical trigger. Complementing this, Trigger-Aware Quantization (TAQ) applies trigger-aware mixed precision in cross-attention, preserving trigger-conditioned key/value rows (and their attention weights) while aggressively quantizing the remaining pathways for memory-efficient inference. Experiments show that PersonalQ improves intent alignment over retrieval and reranking baselines, while TAQ consistently offers a stronger compression-quality trade-off than prior diffusion PTQ methods, enabling scalable serving of personalized checkpoints without sacrificing fidelity.

MSR-HuBERT: Self-supervised Pre-training for Adaptation to Multiple Sampling Rates

cs.AI, q-bio.NC updates on arXiv.org

By: Zikang Huang · Meng Ge · Tianrui Wang · Xuanchen Li · Xiaobao Wang · Longbiao Wang · Jianwu Dang

25 March 2026 at 12:00

arXiv:2603.23048v1 Announce Type: cross Abstract: Self-supervised learning (SSL) has advanced speech processing. However, existing speech SSL methods typically assume a single sampling rate and struggle with mixed-rate data due to temporal resolution mismatch. To address this limitation, we propose MSRHuBERT, a multi-sampling-rate adaptive pre-training method. Building on HuBERT, we replace its single-rate downsampling CNN with a multi-sampling-rate adaptive downsampling CNN that maps raw waveforms from different sampling rates to a shared temporal resolution without resampling. This design enables unified mixed-rate pre-training and fine-tuning. In experiments spanning 16 to 48 kHz, MSRHuBERT outperforms HuBERT on speech recognition and full-band speech reconstruction, preserving high-frequency detail while modeling low-frequency semantic structure. Moreover, MSRHuBERT retains HuBERT's mask-prediction objective and Transformer encoder, so existing analyses and improvements that were developed for HuBERT can apply directly.

AI Model Modulation with Logits Redistribution

cs.AI, q-bio.NC updates on arXiv.org

By: Zihan Wang · Zhongkui Ma · Xinguo Feng · Zhiyang Mei · Ethan Ma · Derui Wang · Minhui Xue · Guangdong Bai

16 March 2026 at 12:00

arXiv:2603.12755v1 Announce Type: new Abstract: Large-scale models are typically adapted to meet the diverse requirements of model owners and users. However, maintaining multiple specialized versions of the model is inefficient. In response, we propose AIM, a novel model modulation paradigm that enables a single model to exhibit diverse behaviors to meet the specific end requirements. AIM enables two key modulation modes: utility and focus modulations. The former provides model owners with dynamic control over output quality to deliver varying utility levels, and the latter offers users precise control to shift model's focused input features. AIM introduces a logits redistribution strategy that operates in a training data-agnostic and retraining-free manner. We establish a formal foundation to ensure AIM's regulation capability, based on the statistical properties of logits ordering via joint probability distributions. Our evaluation confirms AIM's practicality and versatility for Al model modulation, with tasks spanning image classification, semantic segmentation and text generation, and prevalent architectures including ResNet, SegFormer and Llama.

MedFeat: Model-Aware and Explainability-Driven Feature Engineering with LLMs for Clinical Tabular Prediction

cs.AI, q-bio.NC updates on arXiv.org

By: Zizheng Zhang · Yiming Li · Justin Xu · Jinyu Wang · Rui Wang · Lei Song · Jiang Bian · David W Eyre · Jingjing Fu

4 March 2026 at 13:00

arXiv:2603.02221v1 Announce Type: cross Abstract: In healthcare tabular predictions, classical models with feature engineering often outperform neural approaches. Recent advances in Large Language Models enable the integration of domain knowledge into feature engineering, offering a promising direction. However, existing approaches typically rely on a broad search over predefined transformations, overlooking downstream model characteristics and feature importance signals. We present MedFeat, a feedback-driven and model-aware feature engineering framework that leverages LLM reasoning with domain knowledge and provides feature explanations based on SHAP values while tracking successful and failed proposals to guide feature discovery. By incorporating model awareness, MedFeat prioritizes informative signals that are difficult for the downstream model to learn directly due to its characteristics. Across a broad range of clinical prediction tasks, MedFeat achieves stable improvements over various baselines and discovers clinically meaningful features that generalize under distribution shift, demonstrating robustness across years and from ICU cohorts to general hospitalized patients, thereby offering insights into real-world deployment. Code required to reproduce our experiments will be released, subject to dataset agreements and institutional policies.

From Complex Dynamics to DynFormer: Rethinking Transformers for PDEs

cs.AI, q-bio.NC updates on arXiv.org

By: Pengyu Lai · Yixiao Chen · Dewu Yang · Rui Wang · Feng Wang · Hui Xu

4 March 2026 at 13:00

arXiv:2603.03112v1 Announce Type: cross Abstract: Partial differential equations (PDEs) are fundamental for modeling complex physical systems, yet classical numerical solvers face prohibitive computational costs in high-dimensional and multi-scale regimes. While Transformer-based neural operators have emerged as powerful data-driven alternatives, they conventionally treat all discretized spatial points as uniform, independent tokens. This monolithic approach ignores the intrinsic scale separation of physical fields, applying computationally prohibitive global attention that redundantly mixes smooth large-scale dynamics with high-frequency fluctuations. Rethinking Transformers through the lens of complex dynamics, we propose DynFormer, a novel dynamics-informed neural operator. Rather than applying a uniform attention mechanism across all scales, DynFormer explicitly assigns specialized network modules to distinct physical scales. It leverages a Spectral Embedding to isolate low-frequency modes, enabling a Kronecker-structured attention mechanism to efficiently capture large-scale global interactions with reduced complexity. Concurrently, we introduce a Local-Global-Mixing transformation. This module utilizes nonlinear multiplicative frequency mixing to implicitly reconstruct the small-scale, fast-varying turbulent cascades that are slaved to the macroscopic state, without incurring the cost of global attention. Integrating these modules into a hybrid evolutionary architecture ensures robust long-term temporal stability. Extensive memory-aligned evaluations across four PDE benchmarks demonstrate that DynFormer achieves up to a 95% reduction in relative error compared to state-of-the-art baselines, while significantly reducing GPU memory consumption. Our results establish that embedding first-principles physical dynamics into Transformer architectures yields a highly scalable, theoretically grounded blueprint for PDE surrogate modeling.

Foundation Models in Autonomous Driving: A Survey on Scenario Generation and Scenario Analysis

cs.AI, q-bio.NC updates on arXiv.org

By: Yuan Gao · Mattia Piccinini · Yuchen Zhang · Dingrui Wang · Korbinian Moller · Roberto Brusnicki · Baha Zarrouki · Alessio Gambi · Jan Frederik Totz · Kai Storms · Steven Peters · Andrea Stocco · Bassam Alrifaee · Marco Pavone · Johannes Betz

17 February 2026 at 13:00

arXiv:2506.11526v4 Announce Type: replace-cross Abstract: For autonomous vehicles, safe navigation in complex environments depends on handling a broad range of diverse and rare driving scenarios. Simulation- and scenario-based testing have emerged as key approaches to development and validation of autonomous driving systems. Traditional scenario generation relies on rule-based systems, knowledge-driven models, and data-driven synthesis, often producing limited diversity and unrealistic safety-critical cases. With the emergence of foundation models, which represent a new generation of pre-trained, general-purpose AI models, developers can process heterogeneous inputs (e.g., natural language, sensor data, HD maps, and control actions), enabling the synthesis and interpretation of complex driving scenarios. In this paper, we conduct a survey about the application of foundation models for scenario generation and scenario analysis in autonomous driving (as of May 2025). Our survey presents a unified taxonomy that includes large language models, vision-language models, multimodal large language models, diffusion models, and world models for the generation and analysis of autonomous driving scenarios. In addition, we review the methodologies, open-source datasets, simulation platforms, and benchmark challenges, and we examine the evaluation metrics tailored explicitly to scenario generation and analysis. Finally, the survey concludes by highlighting the open challenges and research questions, and outlining promising future research directions. All reviewed papers are listed in a continuously maintained repository, which contains supplementary materials and is available at https://github.com/TUM-AVS/FM-for-Scenario-Generation-Analysis.

Reading view