A clinically-oriented foundation model for intraoperative pathology

By: Zihan Zhao · Fengtao Zhou · Ronggang Li · Bing Chu · Xinke Zhang · Xueyi Zheng · Ke Zheng · Xiaobo Wen · Jiabo Ma · Yihui Wang · Jiewei Chen · Chengyou Zheng · Jiangyu Zhang · Yongqin Wen · Jiajia Meng · Ziqi Zeng · Xiaoqing Li · Jing Li · Dan Xie · Yaping Ye · Yu Wang · Hao Chen · Muyan Cai

10 September 2026 at 08:00

Nature Medicine, Published online: 10 September 2026; doi:10.1038/s41591-026-04703-0

CRISP, a vision-based pathology foundation model developed exclusively from frozen section slides, supports treatment decision-making throughout the surgical workflow with superior performance to current foundation models and extensive validation, including in a prospective cohort.

From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems

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

By: Ruizhe Zhou · Xiaoyang Liu · Gaoyuan Du · Yi Zheng · Shouxi Ren · Deepayan Chakrabarti · Dengdu Jiang

26 May 2026 at 12:00

arXiv:2605.23955v1 Announce Type: new Abstract: Deploying machine learning in regulated financial environments -- credit risk, fraud detection, and anti-money laundering -- exposes critical vulnerabilities in algorithmic reproducibility. While early financial ML addressed statistical challenges such as backtest overfitting, deep neural networks and Generative AI have introduced mechanical nondeterminism rooted in hardware and architecture. This survey provides a systems perspective on reproducibility failures across three modalities now dominant in financial AI: tabular models (post-hoc explanation variance), graph networks (stochastic sampling and temporal asynchrony), and LLM-based agentic workflows (batch-dependent divergence and trajectory drift). We supplement the literature analysis with first-party experiments on public financial datasets -- quantifying explanation rank instability in credit scoring, prediction flip rates in GNN-based fraud detection, and tensor-parallel-induced output divergence in LLM entity extraction. We propose a layered evaluation framework linking modality-specific metrics (RBO, D_cos, TDI, PSD) to audit readiness, and empirically validate the complementarity of logit-level and semantic-level determinism measures.

Ginkgolic Acids Degradation by the <em>Ginkgo biloba</em> Endophytic Fungus <em>Fusarium</em> sp. DLT-118

Omics In Lung

By: Lu-Ting Dai · Zhi-Fang Yu · You-Xing Zhao · Yi Zheng

14 April 2026 at 18:00

Foods. 2026 Apr 6;15(7):1247. doi: 10.3390/foods15071247.

ABSTRACT

Ginkgolic acids (GAs), the principal toxic constituents in Ginkgo biloba, pose health risks including cytotoxicity, allergenicity, and pro-inflammatory effects, limiting the application of Ginkgo resources in the food and health product industries. Developing efficient and environmentally friendly removal methods is essential. The endophytic fungus Fusarium sp. DLT-118, isolated from Ginkgo biloba, degraded 96.47% of GAs in Ginkgo biloba leaf extract (GE) at an initial concentration of 1 mg/mL within 7 days at 28 °C, while concurrently enhancing the antioxidant activity of GE, as indicated by a reduction in the 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging IC₅₀ from 755.7 μg/mL to 544.6 μg/mL. Morphological and oxidative stress analyses showed critical cellular adaptations and stress responses under degradation conditions. Integrated multi-omics analysis indicated that GE stress induced the remodeling of fungal amino acid, lipid, and energy metabolism, as well as the adjustment of membrane and transport functions, to facilitate GAs detoxification. Cytotoxicity assays indicated no significant cytotoxicity of the degradation products towards human normal lung epithelial cells (Beas-2B) and gastric mucosal epithelial cells (GES-1). These findings highlight Fusarium sp. DLT-118 as a promising agent for the efficient removal of GAs, offering a potential strategy for the production of GA-reduced Ginkgo-based food and health products.

PMID:41976541 | PMC:PMC13073844 | DOI:10.3390/foods15071247

UBTF-HSP90A-MIF stress circuit drives lenvatinib resistance and immune exclusion in hepatocellular carcinoma

Omics in Hepatocellular

By: Shiping Chen · Biao Wang · Yanfei Wu · Zongjuan Li · Qiuyi Zheng · Zhigang Fu · Weifeng Hong · Shisuo Du · Yang Zhang

7 April 2026 at 18:00

J Adv Res. 2026 Apr 5:S2090-1232(26)00280-8. doi: 10.1016/j.jare.2026.04.002. Online ahead of print.

ABSTRACT

INTRODUCTION: The clinical benefit of combining lenvatinib with PD-1 blockade in HCC is frequently constrained by adaptive resistance and the development of an immune-cold tumor microenvironment.

OBJECTIVES: This study aimed to elucidate the molecular mechanisms underlying adaptive resistance and immune exclusion during lenvatinib-PD-1 therapy in HCC, with a particular focus on a UBTF/HSP90A/MIF regulatory circuit. We examined whether genetic or pharmacologic targeting of macrophage migration inhibitory factor (MIF) could restore lenvatinib sensitivity, remodel the tumor immune microenvironment, and serve as a predictive biomarker in clinical cohorts.

METHODS: Paired lenvatinib-sensitive and -resistant HCC models were interrogated using integrated multi-omic and functional approaches, including RNA sequencing, promoter pull-down assays, ChIP, luciferase reporter assays, PLA, and flow cytometry. Key findings were validated in patient-derived organoids and xenografts, as well as in an immunocompetent hydrodynamic HCC mouse model. Clinical relevance was evaluated in independent cohorts treated with lenvatinib plus anti-PD-1 therapy.

RESULTS: UBTF directly bound to and transcriptionally activated the HSP90A promoter, resulting in increased HSP90A expression and stabilization of MIF. MIF signaling through CD74 co-activated the PI3K-AKT and MAPK pathways, sustaining tumor cell proliferation under lenvatinib pressure. Single-cell RNA sequencing and multiplex immunohistochemistry revealed macrophage enrichment and CD8⁺ T-cell exclusion in resistant tumors. Genetic ablation of Mif (Alb-Cre; Mif^flox/flox) or pharmacologic inhibition with 4-IPP (4-Iodo-6-phenylpyrimidine) restored lenvatinib sensitivity, reprogrammed the tumor immune microenvironment, and, when combined with PD-1 blockade, achieved superior tumor control and prolonged survival. In clinical datasets, low pretreatment MIF expression was associated with improved responses to lenvatinib plus PD-1 therapy.

CONCLUSIONS: These findings define a UBTF/HSP90A/MIF axis linking proteostasis and cytokine signaling to immune-metabolic dysfunction and lenvatinib resistance in HCC. MIF emerges as both a mechanistic driver and a predictive biomarker, supporting prospective evaluation of therapeutic strategies combining lenvatinib-PD-1 with MIF- or HSP90A-targeted interventions to personalize TKI-ICI therapy.

PMID:41946392 | DOI:10.1016/j.jare.2026.04.002

InfoTok: Information-Theoretic Regularization for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs

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

By: Lv Tang · Tianyi Zheng · Bo Li · Xingyu Li

7 April 2026 at 12:00

arXiv:2602.01554v2 Announce Type: replace-cross Abstract: Unified multimodal large language models (MLLMs) aim to unify image understanding and image generation within a single framework, where a shared visual tokenizer serves as the sole interface that maps high-dimensional images into a limited token budget for downstream multimodal reasoning and synthesis. However, existing shared-token designs are largely architecture-driven and lack an explicit criterion for what information should be preserved to simultaneously support semantic abstraction and visual detail. In this paper, we adopt a capacity-constrained perspective, viewing the shared tokenizer as a compute-bounded learner whose finite representational budget should prioritize reusable structure over hard-to-exploit high-entropy variations and redundancy. Motivated by this view, we propose \textbf{\textit{InfoTok}}, an information-regularized tokenization mechanism grounded in the Information Bottleneck (IB) principle. InfoTok explicitly controls information flow from images to shared tokens to multimodal outputs by imposing mutual-information (MI) constraints that enforce a principled trade-off between compression and task relevance, while also encouraging cross-modal consistency. Because MI is intractable for high-dimensional visual representations, we instantiate InfoTok with practical, differentiable dependence estimators, including a variational IB formulation and a Hilbert Schmidt Independence Criterion (HSIC) based alternative. Integrated into three representative unified MLLMs without introducing any additional training data, InfoTok consistently improves both image understanding and generation performance. These results support information-regularized visual tokenization as a sound basis for token learning in unified MLLMs.

Value Flows

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

By: Perry Dong · Chongyi Zheng · Chelsea Finn · Dorsa Sadigh · Benjamin Eysenbach

5 March 2026 at 13:00

arXiv:2510.07650v3 Announce Type: replace-cross Abstract: While most reinforcement learning methods today flatten the distribution of future returns to a single scalar value, distributional RL methods exploit the return distribution to provide stronger learning signals and to enable applications in exploration and safe RL. While the predominant method for estimating the return distribution is by modeling it as a categorical distribution over discrete bins or estimating a finite number of quantiles, such approaches leave unanswered questions about the fine-grained structure of the return distribution and about how to distinguish states with high return uncertainty for decision-making. The key idea in this paper is to use modern, flexible flow-based models to estimate the full future return distributions and identify those states with high return variance. We do so by formulating a new flow-matching objective that generates probability density paths satisfying the distributional Bellman equation. Building upon the learned flow models, we estimate the return uncertainty of distinct states using a new flow derivative ODE. We additionally use this uncertainty information to prioritize learning a more accurate return estimation on certain transitions. We compare our method (Value Flows) with prior methods in the offline and online-to-online settings. Experiments on $37$ state-based and $25$ image-based benchmark tasks demonstrate that Value Flows achieves a $1.3\times$ improvement on average in success rates. Website: https://pd-perry.github.io/value-flows Code: https://github.com/chongyi-zheng/value-flows

Reading view