❌

Normal view

Received β€” 2 April 2026 ⏭

PromptForge-350k: A Large-Scale Dataset and Contrastive Framework for Prompt-Based AI Image Forgery Localization

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Jianpeng Wang Β· Haoyu Wang Β· Baoying Chen Β· Jishen Zeng Β· Yiming Qin Β· Yiqi Yang Β· Zhongjie Ba
1 April 2026 at 12:00
arXiv:2603.29386v1 Announce Type: cross Abstract: The rapid democratization of prompt-based AI image editing has recently exacerbated the risks associated with malicious content fabrication and misinformation. However, forgery localization methods targeting these emerging editing techniques remain significantly under-explored. To bridge this gap, we first introduce a fully automated mask annotating framework that leverages keypoint alignment and semantic space similarity to generate precise ground-truth masks for edited regions. Based on this framework, we construct PromptForge-350k, a large-scale forgery localization dataset covering four state-of-the-art prompt-based AI image editing models, thereby mitigating the data scarcity in this domain. Furthermore, we propose ICL-Net, an effective forgery localization network featuring a triple-stream backbone and intra-image contrastive learning. This design enables the model to capture highly robust and generalizable forensic features. Extensive experiments demonstrate that our method achieves an IoU of 62.5% on PromptForge-350k, outperforming SOTA methods by 5.1%. Additionally, it exhibits strong robustness against common degradations with an IoU drop of less than 1%, and shows promising generalization capabilities on unseen editing models, achieving an average IoU of 41.5%.
  • βœ‡Omics In Lung
  • Tumorigenesis and Tumor Microenvironment in Lung Cancer Puneet Dhillon Β· Moshe Carroll Β· Haiying Cheng
    Curr Issues Mol Biol. 2026 Feb 26;48(3):247. doi: 10.3390/cimb48030247.ABSTRACTLung cancer remains a leading cause of cancer mortality worldwide and continues to impose substantial clinical and economic burdens. Beyond tumor-intrinsic oncogenic drivers, disease progression and therapy response are shaped by the tumor microenvironment (TME), including immune cells, cancer-associated fibroblasts (CAFs), endothelial cells, extracellular matrix, inflammatory mediators, etc. In lung cancer, chronic i
     
  • 🏷️
  • ↗️

Tumorigenesis and Tumor Microenvironment in Lung Cancer

βœ‡Omics In Lung
By: Puneet Dhillon Β· Moshe Carroll Β· Haiying Cheng
28 March 2026 at 18:00

Curr Issues Mol Biol. 2026 Feb 26;48(3):247. doi: 10.3390/cimb48030247.

ABSTRACT

Lung cancer remains a leading cause of cancer mortality worldwide and continues to impose substantial clinical and economic burdens. Beyond tumor-intrinsic oncogenic drivers, disease progression and therapy response are shaped by the tumor microenvironment (TME), including immune cells, cancer-associated fibroblasts (CAFs), endothelial cells, extracellular matrix, inflammatory mediators, etc. In lung cancer, chronic injury from tobacco smoke, airway disease, and treatment itself remodels local tissue programs that can either support antitumor immunity or promote immune exclusion, fibrosis, and metastatic seeding. Here, we analyze recent evidence linking lung tumorigenesis to TME ecology across histologies, with emphasis on CAF heterogeneity, spatial organization of immune niches, and the distinct microenvironments that govern organ-specific metastasis (including brain metastasis). We also evaluate emerging therapeutic strategies that aim to target or reprogram the TME, including perioperative immune checkpoint blockade, combined immunotherapy-radiotherapy approaches, and pathways such as IL-6 and TGF-Ξ² that coordinate immune suppression and stromal remodeling. Finally, we outline key gaps and potential future directions, such as longitudinal and spatial multi-omics, better biomarkers of stromal state, and trial designs that account for dynamic microenvironmental adaptation.

PMID:41899399 | PMC:PMC13025777 | DOI:10.3390/cimb48030247

There are no more articles
❌