Post Content
Pfizer’s Lyme disease vaccine trial hampered by lack of cases
Pfizer’s final-stage trial of its long-awaited Lyme disease vaccine missed its primary endpoint, but the drugmaker said it still plans to try to take the experimental shot to regulators for approval. The company said there …
Dizal breaks new ground for EGFR drugs in lung cancer
Dizal’s Zegfrovy has taken the oral EGFR inhibitor class into new territory, with a phase 3 win as a frontline treatment for a form of lung cancer.
Earendil raises $787m as lead TL1A antibody starts phase 2
Earendil has raised an impressive $787m in financing for its AI-powered drug platform, attracting yet more buy-in from Sanofi.
The Bay Area’s animal welfare movement wants to recruit AI
In early February, animal welfare advocates and AI researchers gathered in stocking feet at Mox, a scrappy, shoes-free coworking space in San Francisco. Yellow and red canopies billowed overhead, Persian rugs blanketed the floor, and mosaic lamps glowed beside potted plants. In the common area, a wildlife advocate spoke passionately to a crowd lounging in […]
Implementing AI innovation in radiology departments in the English NHS: a qualitative study on the experiences of professionals, patient groups and innovators
IntroductionDigital solutions and Artificial Intelligence (AI) innovations are often presented as the answer to many challenges faced by healthcare systems around the world. The UK government has made significant investments in this area, yet there have been concerns about the challenges faced when these technologies are implemented in practice. The aim of this study was […]
Co-creating a program theory and evaluability assessment for an Irish single-session, synchronous chat-based youth mental health intervention: implications for outcome evaluation
IntroductionSingle-session online synchronous chat offers immediate, anonymous, single-session support for young people. However, the drop-in format attracts a diverse population with urgent and varied needs, creating challenges for evaluation. Standardized outcome measures may not capture short-term changes, and randomized controlled trials may be ethically inappropriate. These constraints point to the value of theory-based evaluation approaches […]
LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
arXiv:2603.19312v1 Announce Type: cross Abstract: Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces, yet existing methods remain fragile, relying on complex multi-term losses, exponential moving averages, pre-trained encoders, or auxiliary supervision to avoid representation collapse. In this work, we introduce LeWorldModel (LeWM), the first JEPA that […]
Do Post-Training Algorithms Actually Differ? A Controlled Study Across Model Scales Uncovers Scale-Dependent Ranking Inversions
arXiv:2603.19335v1 Announce Type: cross Abstract: Post-training alignment has produced dozens of competing algorithms — DPO, SimPO, KTO, GRPO, and others — yet practitioners lack controlled comparisons to guide algorithm selection. We present OXRL, a unified framework implementing 51 post-training algorithms with identical infrastructure, enabling the first large-scale apples-to-apples evaluation. Our study spans 8 algorithms across […]
Exploring Subnetwork Interactions in Heterogeneous Brain Network via Prior-Informed Graph Learning
arXiv:2603.19307v1 Announce Type: cross Abstract: Modeling the complex interactions among functional subnetworks is crucial for the diagnosis of mental disorders and the identification of functional pathways. However, learning the interactions of the underlying subnetworks remains a significant challenge for existing Transformer-based methods due to the limited number of training samples. To address these challenges, we […]
HiPath: Hierarchical Vision-Language Alignment for Structured Pathology Report Prediction
arXiv:2603.19957v1 Announce Type: cross Abstract: Pathology reports are structured, multi-granular documents encoding diagnostic conclusions, histological grades, and ancillary test results across one or more anatomical sites; yet existing pathology vision-language models (VLMs) reduce this output to a flat label or free-form text. We present HiPath, a lightweight VLM framework built on frozen UNI2 and Qwen3 […]
Auditing Google’s AI Overviews and Featured Snippets: A Case Study on Baby Care and Pregnancy
arXiv:2511.12920v3 Announce Type: replace-cross Abstract: Google Search increasingly surfaces AI-generated content through features like AI Overviews (AIO) and Featured Snippets (FS), which users frequently rely on despite having no control over their presentation. Through a systematic algorithm audit of 1,508 real baby care and pregnancy-related queries, we evaluate the quality and consistency of these information […]