❌

Normal view

Received — 2 April 2026 ⏭

Tracking vs. Deciding: The Dual-Capability Bottleneck in Searchless Chess Transformers

✇cs.AI, q-bio.NC updates on arXiv.org
By: Quanhao Li · Wei Jiang
1 April 2026 at 12:00
arXiv:2603.29761v1 Announce Type: new Abstract: A human-like chess engine should mimic the style, errors, and consistency of a strong human player rather than maximize playing strength. We show that training from move sequences alone forces a model to learn two capabilities: state tracking, which reconstructs the board from move history, and decision quality, which selects good moves from that reconstructed state. These impose contradictory data requirements: low-rated games provide the diversity needed for tracking, while high-rated games provide the quality signal for decision learning. Removing low-rated data degrades performance. We formalize this tension as a dual-capability bottleneck, P

IMPASTO: Integrating Model-Based Planning with Learned Dynamics Models for Robotic Oil Painting Reproduction

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yingke Wang · Hao Li · Yifeng Zhu · Hong-Xing Yu · Ken Goldberg · Li Fei-Fei · Jiajun Wu · Yunzhu Li · Ruohan Zhang
1 April 2026 at 12:00
arXiv:2603.29315v1 Announce Type: cross Abstract: Robotic reproduction of oil paintings using soft brushes and pigments requires force-sensitive control of deformable tools, prediction of brushstroke effects, and multi-step stroke planning, often without human step-by-step demonstrations or faithful simulators. Given only a sequence of target oil painting images, can a robot infer and execute the stroke trajectories, forces, and colors needed to reproduce it? We present IMPASTO, a robotic oil-painting system that integrates learned pixel dynamics models with model-based planning. The dynamics models predict canvas updates from image observations and parameterized stroke actions; a receding-horizon model predictive control optimizer then plans trajectories and forces, while a force-sensitive controller executes strokes on a 7-DoF robot arm. IMPASTO integrates low-level force control, learned dynamics models, and high-level closed-loop planning, learns solely from robot self-play, and approximates human artists' single-stroke datasets and multi-stroke artworks, outperforming baselines in reproduction accuracy. Project website: https://impasto-robopainting.github.io/

MacTok: Robust Continuous Tokenization for Image Generation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hengyu Zeng · Xin Gao · Guanghao Li · Yuxiang Yan · Jiaoyang Ruan · Junpeng Ma · Haoyu Albert Wang · Jian Pu
1 April 2026 at 12:00
arXiv:2603.29634v1 Announce Type: cross Abstract: Continuous image tokenizers enable efficient visual generation, and those based on variational frameworks can learn smooth, structured latent representations through KL regularization. Yet this often leads to posterior collapse when using fewer tokens, where the encoder fails to encode informative features into the compressed latent space. To address this, we introduce \textbf{MacTok}, a \textbf{M}asked \textbf{A}ugmenting 1D \textbf{C}ontinuous \textbf{Tok}enizer that leverages image masking and representation alignment to prevent collapse while learning compact and robust representations. MacTok applies both random masking to regularize latent learning and DINO-guided semantic masking to emphasize informative regions in images, forcing the model to encode robust semantics from incomplete visual evidence. Combined with global and local representation alignment, MacTok preserves rich discriminative information in a highly compressed 1D latent space, requiring only 64 or 128 tokens. On ImageNet, MacTok achieves a competitive gFID of 1.44 at 256$\times$256 and a state-of-the-art 1.52 at 512$\times$512 with SiT-XL, while reducing token usage by up to 64$\times$. These results confirm that masking and semantic guidance together prevent posterior collapse and achieve efficient, high-fidelity tokenization.

Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jingwei Ni · Yihao Liu · Xinpeng Liu · Yutao Sun · Mengyu Zhou · Pengyu Cheng · Dexin Wang · Erchao Zhao · Xiaoxi Jiang · Guanjun Jiang
1 April 2026 at 12:00
arXiv:2603.25158v3 Announce Type: replace Abstract: Equipping Large Language Model (LLM) agents with domain-specific skills is critical for tackling complex tasks. Yet, manual authoring creates a severe scalability bottleneck. Conversely, automated skill generation often yields fragile or fragmented results because it either relies on shallow parametric knowledge or sequentially overfits to non-generalizable trajectory-local lessons. To overcome this, we introduce Trace2Skill, a framework that mirrors how human experts author skills: by holistically analyzing broad execution experience before distilling it into a single, comprehensive guide. Instead of reacting sequentially to individual trajectories, Trace2Skill dispatches a parallel fleet of sub-agents to analyze a diverse pool of executions. It extracts trajectory-specific lessons and hierarchically consolidates them into a unified, conflict-free skill directory via inductive reasoning. Trace2Skill supports both deepening existing human-written skills and creating new ones from scratch. Experiments in challenging domains, such as spreadsheet, VisionQA and math reasoning, show that Trace2Skill significantly improves upon strong baselines, including Anthropic's official xlsx skills. Crucially, this trajectory-grounded evolution does not merely memorize task instances or model-specific quirks: evolved skills transfer across LLM scales and generalize to OOD settings. For example, skills evolved by Qwen3.5-35B on its own trajectories improved a Qwen3.5-122B agent by up to 57.65 absolute percentage points on WikiTableQuestions. Ultimately, our results demonstrate that complex agent experience can be packaged into highly transferable, declarative skills -- requiring no parameter updates, no external retrieval modules, and utilizing open-source models as small as 35B parameters.

Incorporating LLM Embeddings for Variation Across the Human Genome

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hongqian Niu · Jordan Bryan · Jacob Williams · Hufeng Zhou · Haoyu Zhang · Xihao Li · Didong Li
1 April 2026 at 12:00
arXiv:2509.20702v2 Announce Type: replace-cross Abstract: Recent advances in large language model (LLM) embeddings have enabled powerful representations for biological data, but most applications to date focus on gene-level information. We present one of the first systematic frameworks to generate genetic variant-level embeddings across the entire human genome. Using curated annotations from FAVOR, ClinVar, and the GWAS Catalog, we construct functional text descriptions for 8.9 billion possible variants and generated embeddings at three scales: 1.5 million HapMap3/MEGA variants, 90 million imputed UK Biobank (UKB) variants, and 9 billion all possible variants. Embeddings were produced using general purpose models including both OpenAI's text-embedding-3-large and the open-source Qwen3-Embedding-0.6B models. Baseline quality control experiments demonstrate high predictive accuracy for variant-level properties, validating the embeddings as structured representations of genomic variation. We further apply them to real-world embedding-augmented genetic risk predictions that demonstrate the performance of using LLM embeddings in polygenic risk score (PRS) style predictions over the UK Biobank cohort data. These resources, publicly available on Hugging Face, provide a foundation for advancing large-scale genomic discovery and precision medicine.

SecureVibeBench: Evaluating Secure Coding Capabilities of Code Agents with Realistic Vulnerability Scenarios

✇cs.AI, q-bio.NC updates on arXiv.org
By: Junkai Chen · Huihui Huang · Yunbo Lyu · Junwen An · Jieke Shi · Chengran Yang · Ting Zhang · Haoye Tian · Yikun Li · Zhenhao Li · Xin Zhou · Xing Hu · David Lo
1 April 2026 at 12:00
arXiv:2509.22097v2 Announce Type: replace-cross Abstract: Large language model-powered code agents are rapidly transforming software engineering, yet the security risks of their generated code have become a critical concern. Existing benchmarks have provided valuable insights, but they fail to capture scenarios in which vulnerabilities are actually introduced by human developers, making fair comparisons between humans and agents infeasible. We therefore introduce SecureVibeBench, a benchmark of 105 C/C++ secure coding tasks sourced from 41 projects in OSS-Fuzz for code agents. SecureVibeBench has the following features: (i) realistic task settings that require multi-file edits in large repositories, (ii)~aligned contexts based on real-world open-source vulnerabilities with precisely identified vulnerability introduction points, and (iii) comprehensive evaluation that combines functionality testing and security checking with both static and dynamic oracles. We evaluate 5 popular code agents like OpenHands, supported by 5 LLMs (e.g., Claude sonnet 4.5) on SecureVibeBench. Results show that current agents struggle to produce both correct and secure code, as even the best-performing one, produces merely 23.8\% correct and secure solutions on SecureVibeBench.

Heracles: Bridging Precise Tracking and Generative Synthesis for General Humanoid Control

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zelin Tao · Zeran Su · Peiran Liu · Jingkai Sun · Wenqiang Que · Jiahao Ma · Jialin Yu · Jiahang Cao · Pihai Sun · Hao Liang · Gang Han · Wen Zhao · Zhiyuan Xu · Jian Tang · Qiang Zhang · Yijie Guo
1 April 2026 at 12:00
arXiv:2603.27756v2 Announce Type: replace-cross Abstract: Achieving general-purpose humanoid control requires a delicate balance between the precise execution of commanded motions and the flexible, anthropomorphic adaptability needed to recover from unpredictable environmental perturbations. Current general controllers predominantly formulate motion control as a rigid reference-tracking problem. While effective in nominal conditions, these trackers often exhibit brittle, non-anthropomorphic failure modes under severe disturbances, lacking the generative adaptability inherent to human motor control. To overcome this limitation, we propose Heracles, a novel state-conditioned diffusion middleware that bridges precise motion tracking and generative synthesis. Rather than relying on rigid tracking paradigms or complex explicit mode-switching, Heracles operates as an intermediary layer between high-level reference motions and low-level physics trackers. By conditioning on the robot's real-time state, the diffusion model implicitly adapts its behavior: it approximates an identity map when the state closely aligns with the reference, preserving zero-shot tracking fidelity. Conversely, when encountering significant state deviations, it seamlessly transitions into a generative synthesizer to produce natural, anthropomorphic recovery trajectories. Our framework demonstrates that integrating generative priors into the control loop not only significantly enhances robustness against extreme perturbations but also elevates humanoid control from a rigid tracking paradigm to an open-ended, generative general-purpose architecture.

Targeting FTO shows therapeutic potential in esophageal squamous cell carcinoma by modulating microRNA biogenesis

✇Oncogene - Issue - nature.com science feeds
By: Wei Zhou · Chunyan Wang · Changhao Li · Pengxiang Chen · Yuchen Liu · Hongyuan Mao · Pengfei Zhu · Yuxiao Gong · Yufeng Cheng · Lin Zhang
1 April 2026 at 08:00

Oncogene, Published online: 01 April 2026; doi:10.1038/s41388-026-03754-4

Targeting FTO shows therapeutic potential in esophageal squamous cell carcinoma by modulating microRNA biogenesis

Doxorubicin promotes the production of inflammatory cytokines in tumor-associated macrophages through activating lactate dehydrogenase A

✇Cell Death Discovery nature.com science feeds
By: Bin Liu · Wei Yang · Shuo Feng · Xingning Jiang · Jingfan Lu · Fei Wang · Yaowen Hu · Yanhao Liu · Haifeng Ma · Ting Sun
31 March 2026 at 08:00

Cell Death Discovery, Published online: 31 March 2026; doi:10.1038/s41420-026-03014-0

Doxorubicin promotes the production of inflammatory cytokines in tumor-associated macrophages through activating lactate dehydrogenase A

Multi-omics analysis reveals AR as a potential prognostic factor and immune-related therapeutic target in gastric cancer

✇Omics in Gastric
By: Tianru Ben · Zonghao Liu · Shitong Cheng · Yanlu Bai · Linlin Gao · Jingya Zhang · Ruixuan Wang · Nan Xia
27 March 2026 at 18:00

Biochem Biophys Rep. 2026 Mar 16;46:102537. doi: 10.1016/j.bbrep.2026.102537. eCollection 2026 Jun.

ABSTRACT

BACKGROUND: Although studies have shown that the androgen receptor (AR) is associated with tumor progression and malignant regulation, its role in the tumor immune microenvironment and predictive value for prognosis and immunotherapy response in various cancer types have not been systematically analyzed.

METHODS: In this paper, multi-omics techniques was used to analyze AR comprehensively.

RESULTS: A comprehensive pan-cancer analysis revealed that the AR was expressed in a variety of tumors, especially as a risk factor for poor prognosis in gastric cancer. In addition, gene set enrichment analysis showed that the AR promotes cell proliferation and tumor cell invasion and regulates anti-tumor response. Immune score, immune cell infiltration, and anticancer immune cycle analysis showed that high AR levels were correlated with low infiltration of CD4+ T cells and NKT cells, high infiltration of Th2 cells and MDSCs, negatively correlated with antigen-presenting molecules, and positively correlated with various immune-negative regulatory molecules. Single-cell sequencing highlighted the heterogeneous expression of ARs in different cell types, particularly in epithelial cells, where high AR levels were associated with the enhanced activity of tumor-promoting pathways.

CONCLUSIONS: In conclusion, this study highlights the potential of the AR as a novel biomarker for gastric cancer prognosis and immunotherapy efficacy, expanding its applicability in the development of new antitumor drugs.

PMID:41890218 | PMC:PMC13014673 | DOI:10.1016/j.bbrep.2026.102537

SVNeoPP: A Workflow for Structural-Variant-Derived Neoantigen Prediction and Prioritization Using Multi-Omics Data

✇Omics in Hepatocellular
By: Wanyang An · Xiaoxiu Tan · Zhenhao Liu · Li Zou · Manman Lu · Lu Xie
27 March 2026 at 18:00

Biology (Basel). 2026 Mar 19;15(6):492. doi: 10.3390/biology15060492.

ABSTRACT

BACKGROUND: Tumor neoantigens are key targets for personalized vaccines and T-cell therapies, yet most pipelines focus on neoantigens derived from SNV/small indel and often yield a limited number of high-quality candidates. SVs are prevalent in tumors and can generate novel chimeric sequences and neopeptides, making them a promising additional source of neoantigens. However, SV-derived neoantigen prediction remains challenging due to breakpoint uncertainty, isoform-dependent coding inference, and limited integration of multi-dimensional evidence and reproducibility.

METHODS: We developed SVNeoPP (Structural Variant Neoantigen Prediction and Prioritization), an end-to-end workflow for SV-derived neoantigen analysis. SVNeoPP takes WGS and RNA-seq as inputs, performs SV calling and annotation, and reconstructs altered transcripts and coding sequences in a traceable, isoform-aware manner to generate candidate peptides. Candidates are prescreened by integrating antigen-processing features with HLA binding prediction, and then hierarchically filtered and prioritized based on transcript expression, LC-MS/MS proteomics evidence, immunogenicity predictions, and sequence similarity to experimentally validated neoantigen databases. SVNeoPP is implemented in Snakemake to enable modular extension, checkpoint-based restarts, and end-to-end reproducibility.

RESULTS: Using a hepatocellular carcinoma (HCC) multi-omics dataset as a proof of concept, we demonstrated the performance of SVNeoPP and obtained a high-priority shortlist of candidate peptides. Compared with other methods, SVNeoPP substantially expanded the candidate search space for SV-derived neoantigens and showed more favorable distributions of antigen-processing and HLA binding features.

CONCLUSIONS: SVNeoPP provides a reusable, traceable, and interpretable multi-dimensional evidence-driven framework for SV-derived neoantigens. As a complementary module to SNV/small-indel pipelines, it broadens the neoantigen candidate repertoire and generates ranked candidates with interpretable evidence to facilitate downstream prioritization and decision-making.

PMID:41892252 | PMC:PMC13024079 | DOI:10.3390/biology15060492

Received — 26 March 2026 ⏭

Suicidal Thoughts and Behaviors Among Chinese Adolescents in Relation to Negative Life Events, Internet Addiction, and Sexual Abuse: Cross-Sectional Study

✇Journal of Medical Internet Research
By: Juanfang Zhu · Qinqin Jiang · Feng Zhang · Hao Liu · Shaojie Yu · Jinhai Sun · Zhe Zhao · Lijuan Liu · Lei Yuan
26 March 2026 at 04:30
Background: Increasing suicidal thoughts and behaviors (STB) among adolescents raise social concerns and have a well-recognized association with sexual abuse (SA). However, research regarding the mechanisms explaining the association between SA and STB remains limited. Objective: This study aims to examine the chained mediating effects of negative life events (NLE) and internet addiction (IA) between SA and STB among adolescents in China. Methods: This cross-sectional study used data from the Science Database of the People Mental Health survey conducted between March 2013 and December 2022 by the National Population Health Data Center of the National Research Institute for Family Planning. Through stratified sampling, 20,893 adolescents were recruited from 16 Chinese provinces. After excluding samples with missing relevant variables, 10,664 (55.89%; aged 16-17.9 y; n=5826, 54.63% women) adolescents were included in the final analysis. STB was the outcome variable, with NLE and IA as mediators, all assessed via a questionnaire that was uniformly administered by trained investigators in school settings. The Pearson χ test was used to analyze the association between SA and STB. Using a combination of multiple linear regression and bootstrap testing, the study constructed a chain mediation model to explore how SA influences STB in adolescents through NLE and IA. Results: The scores for SA, NLE, IA, and STB were 1.330 (SD 1.714), 51.960 (SD 23.822), 34.88 (SD 13.852), and 0.690 (SD 1.396), respectively. Multiple linear regression analysis indicated SA was associated with NLE (β=2.382, 95% CI 2.112‐2.653;

UniFluids: Unified Neural Operator Learning with Conditional Flow-matching

✇cs.AI, q-bio.NC updates on arXiv.org
By: Haosen Li · Qi Meng · Jiahao Li · Rui Zhang · Ruihua Song · Liang Ma · Zhi-Ming Ma
25 March 2026 at 12:00
arXiv:2603.22309v1 Announce Type: cross Abstract: Partial differential equation (PDE) simulation holds extensive significance in scientific research. Currently, the integration of deep neural networks to learn solution operators of PDEs has introduced great potential. In this paper, we present UniFluids, a conditional flow-matching framework that harnesses the scalability of diffusion Transformer to unify learning of solution operators across diverse PDEs with varying dimensionality and physical variables. Unlike the autoregressive PDE foundation models, UniFluids adopts flow-matching to achieve parallel sequence generation, making it the first such approach for unified operator learning. Specifically, the introduction of a unified four-dimensional spatiotemporal representation for the heterogeneous PDE datasets enables joint training and conditional encoding. Furthermore, we find the effective dimension of the PDE dataset is much lower than its patch dimension. We thus employ $x$-prediction in the flow-matching operator learning, which is verified to significantly improve prediction accuracy. We conduct a large-scale evaluation of UniFluids on several PDE datasets covering spatial dimensions 1D, 2D and 3D. Experimental results show that UniFluids achieves strong prediction accuracy and demonstrates good scalability and cross-scenario generalization capability. The code will be released later.

URA-Net: Uncertainty-Integrated Anomaly Perception and Restoration Attention Network for Unsupervised Anomaly Detection

✇cs.AI, q-bio.NC updates on arXiv.org
By: Wei Luo · Peng Xing · Yunkang Cao · Haiming Yao · Weiming Shen · Zechao Li
25 March 2026 at 12:00
arXiv:2603.22840v1 Announce Type: cross Abstract: Unsupervised anomaly detection plays a pivotal role in industrial defect inspection and medical image analysis, with most methods relying on the reconstruction framework. However, these methods may suffer from over-generalization, enabling them to reconstruct anomalies well, which leads to poor detection performance. To address this issue, instead of focusing solely on normality reconstruction, we propose an innovative Uncertainty-Integrated Anomaly Perception and Restoration Attention Network (URA-Net), which explicitly restores abnormal patterns to their corresponding normality. First, unlike traditional image reconstruction methods, we utilize a pre-trained convolutional neural network to extract multi-level semantic features as the reconstruction target. To assist the URA-Net learning to restore anomalies, we introduce a novel feature-level artificial anomaly synthesis module to generate anomalous samples for training. Subsequently, a novel uncertainty-integrated anomaly perception module based on Bayesian neural networks is introduced to learn the distributions of anomalous and normal features. This facilitates the estimation of anomalous regions and ambiguous boundaries, laying the foundation for subsequent anomaly restoration. Then, we propose a novel restoration attention mechanism that leverages global normal semantic information to restore detected anomalous regions, thereby obtaining defect-free restored features. Finally, we employ residual maps between input features and restored features for anomaly detection and localization. The comprehensive experimental results on two industrial datasets, MVTec AD and BTAD, along with a medical image dataset, OCT-2017, unequivocally demonstrate the effectiveness and superiority of the proposed method.

Exposed phosphatidylserine is an inhibitory molecule in T cell exhaustion

✇Nature - Issue - nature.com science feeds
By: Christopher B. Medina · Ewelina Sobierajska · Minghao Gong · Daniel T. McManus · Maheshwor Thapa · Judong Lee · Se Jin Im · Jason E. Toombs · Joshua M. Mitchell · Yating Wang · Jennifer W. Carlisle · Gordon J. Freeman · Viraj A. Master · Suresh S. Ramalingam · Haydn T. Kissick · Shuzhao Li · Rolf A. Brekken · Rafi Ahmed
25 March 2026 at 08:00

Nature, Published online: 25 March 2026; doi:10.1038/s41586-026-10266-4

Insights into the mechanism by which phosphatidylserine functions as a non-classical inhibitory molecule during T cell exhaustion, and how phosphatidylserine-targeting antibodies enhance T cell responses are explored.

Spatial Omics in Gastrointestinal Oncology: Recent Advances, Therapeutic Insights, and Clinical Translation

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Lin Zhong · Qian Li · Ting Xiong · Shengzheng Lin · Kaining Wang · Guanying Li · Ye Yan · Jianhao Liu · Chuansong Xue
23 March 2026 at 18:00

J Cancer. 2026 Jan 30;17(3):515-523. doi: 10.7150/jca.127381. eCollection 2026.

ABSTRACT

Gastrointestinal (GI) cancers remain a leading cause of cancer-related morbidity and mortality worldwide, largely due to their molecular heterogeneity, complex tumor microenvironment (TME), and variable treatment responses. In recent years, the emergence of spatially resolved omics technologies-encompassing spatial transcriptomics, proteomics, metabolomics, and epigenomics-has revolutionized the ability to interrogate tumor architecture with unprecedented resolution. These methods enable precise mapping of cellular and molecular interactions within intact tissue contexts, thereby uncovering spatially defined niches that influence tumor progression, immune evasion, and therapeutic resistance. In GI malignancies such as colorectal, gastric, and esophageal cancers, spatial omics have provided critical insights into cancer-stromal-immune crosstalk, identified predictive biomarkers for immunotherapy and targeted agents, and guided the development of novel therapeutic strategies. This review synthesizes the latest advances in spatial omics applied to GI oncology over the past five years, with an emphasis on their integration into early diagnosis, treatment stratification, and real-time monitoring of therapeutic efficacy. We also discuss current challenges, including standardization, data integration, and clinical validation, as well as future directions for incorporating spatial profiling into routine oncology practice. By bridging the gap between bench discoveries and bedside applications, spatial omics hold transformative potential for achieving truly personalized treatment in gastrointestinal cancers.

PMID:41869445 | PMC:PMC13003551 | DOI:10.7150/jca.127381

Received — 17 March 2026 ⏭

Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimization

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xudong Wang · Chaoning Zhang · Jiaquan Zhang · Chenghao Li · Qigan Sun · Sung-Ho Bae · Peng Wang · Ning Xie · Jie Zou · Yang Yang · Hengtao Shen
16 March 2026 at 12:00
arXiv:2603.12933v1 Announce Type: new Abstract: Large Language Model (LLM)-driven Multi-Agent Systems (MAS) have demonstrated strong capability in complex reasoning and tool use, and heterogeneous agent pools further broaden the quality--cost trade-off space. Despite these advances, real-world deployment is often constrained by high inference cost, latency, and limited transparency, which hinders scalable and efficient routing. Existing routing strategies typically rely on expensive LLM-based selectors or static policies, and offer limited controllability for semantic-aware routing under dynamic loads and mixed intents, often resulting in unstable performance and inefficient resource utilization. To address these limitations, we propose AMRO-S, an efficient and interpretable routing framework for Multi-Agent Systems (MAS). AMRO-S models MAS routing as a semantic-conditioned path selection problem, enhancing routing performance through three key mechanisms: First, it leverages a supervised fine-tuned (SFT) small language model for intent inference, providing a low-overhead semantic interface for each query; second, it decomposes routing memory into task-specific pheromone specialists, reducing cross-task interference and optimizing path selection under mixed workloads; finally, it employs a quality-gated asynchronous update mechanism to decouple inference from learning, optimizing routing without increasing latency. Extensive experiments on five public benchmarks and high-concurrency stress tests demonstrate that AMRO-S consistently improves the quality--cost trade-off over strong routing baselines, while providing traceable routing evidence through structured pheromone patterns.

Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yichen Zhang · Da Peng · Zonghao Guo · Zijian Zhang · Xuesong Yang · Tong Sun · Shichu Sun · Yidan Zhang · Yanghao Li · Haiyan Zhao · Wang Xu · Qi Shi · Yangang Sun · Chi Chen · Shuo Wang · Yukun Yan · Xu Han · Qiang Ma · Wei Ke · Liang Wang · Zhiyuan Liu · Maosong Sun
16 March 2026 at 12:00
arXiv:2603.12793v1 Announce Type: cross Abstract: A recent cutting-edge topic in multimodal modeling is to unify visual comprehension and generation within a single model. However, the two tasks demand mismatched decoding regimes and visual representations, making it non-trivial to jointly optimize within a shared feature space. In this work, we present Cheers, a unified multimodal model that decouples patch-level details from semantic representations, thereby stabilizing semantics for multimodal understanding and improving fidelity for image generation via gated detail residuals. Cheers includes three key components: (i) a unified vision tokenizer that encodes and compresses image latent states into semantic tokens for efficient LLM conditioning, (ii) an LLM-based Transformer that unifies autoregressive decoding for text generation and diffusion decoding for image generation, and (iii) a cascaded flow matching head that decodes visual semantics first and then injects semantically gated detail residuals from the vision tokenizer to refine high-frequency content. Experiments on popular benchmarks demonstrate that Cheers matches or surpasses advanced UMMs in both visual understanding and generation. Cheers also achieves 4x token compression, enabling more efficient high-resolution image encoding and generation. Notably, Cheers outperforms the Tar-1.5B on the popular benchmarks GenEval and MMBench, while requiring only 20% of the training cost, indicating effective and efficient (i.e., 4x token compression) unified multimodal modeling. We will release all code and data for future research.

MalURLBench: A Benchmark Evaluating Agents' Vulnerabilities When Processing Web URLs

✇cs.AI, q-bio.NC updates on arXiv.org
By: Dezhang Kong · Zhuxi Wu · Shiqi Liu · Zhicheng Tan · Kuichen Lu · Minghao Li · Qichen Liu · Shengyu Chu · Zhenhua Xu · Xuan Liu · Meng Han
16 March 2026 at 12:00
arXiv:2601.18113v3 Announce Type: replace-cross Abstract: LLM-based web agents have become increasingly popular for their utility in daily life and work. However, they exhibit critical vulnerabilities when processing malicious URLs: accepting a disguised malicious URL enables subsequent access to unsafe webpages, which can cause severe damage to service providers and users. Despite this risk, no benchmark currently targets this emerging threat. To address this gap, we propose MalURLBench, the first benchmark for evaluating LLMs' vulnerabilities to malicious URLs. MalURLBench contains 61,845 attack instances spanning 10 real-world scenarios and 7 categories of real malicious websites. Experiments with 12 popular LLMs reveal that existing models struggle to detect elaborately disguised malicious URLs. We further identify and analyze key factors that impact attack success rates and propose URLGuard, a lightweight defense module. We believe this work will provide a foundational resource for advancing the security of web agents. Our code is available at https://github.com/JiangYingEr/MalURLBench.
Received — 14 March 2026 ⏭

Assembly of helper NLR resistosome clusters upon activation of a coiled-coil NLR

✇Nature - Issue - nature.com science feeds
By: Dongdong Ge · Fausto Andres Ortiz-Morea · Yingpeng Xie · In-Cheol Yeo · Qiaochu Shen · Yulu Zhou · Guangchao Liu · Liang Kong · Libo Shan · Ping He
11 March 2026 at 08:00

Nature, Published online: 11 March 2026; doi:10.1038/s41586-026-10215-1

SUMM2, a coiled-coil NLR, promotes the assembly of higher-order resistosome clusters to initiate cell death in plants.
❌