Research & Innovation

Global: X‑SAM Extends Segmentation Capabilities of Multimodal Large Language ModelsA new multimodal framework called X‑SAM has been introduced to broaden the image‑segmentation abilities of large language models. The research, posted…

Global: Diffusion Language Models Integrate New Knowledge More Efficiently Than Autoregressive ModelsA team of AI researchers announced in an October 2025 arXiv preprint that diffusion‑based large language models (dLLMs) can…

Global: Behavioral Calibration Reduces Hallucinations in LLMsA new training paradigm called behavioral calibration enables large language models to express uncertainty and abstain when confidence is low, markedly decreasing the generation…

Global: Study Questions Embedding-Layer Dominance in LLM Influence Estimation, Highlights Middle Attention Layers and New MetricResearchers have released a new arXiv preprint that reevaluates how training‑sample influence is measured in…

Global: Predictive Model Estimates 180-Day Mortality Risk in Advanced Prostate CancerResearchers have introduced a visit-level model that predicts the probability of death within 180 days for patients with metastatic castration-resistant…

Global: Model Prediction Set Framework Enables Adaptive Online Model Selection for Nonstationary Time SeriesA team of researchers announced a new framework called the Model Prediction Set (MPS) in a paper…

Global: New Bayesian Optimization Framework RAMBO Tackles Multi-Regime ChallengesResearchers have unveiled RAMBO, a Dirichlet Process Mixture of Gaussian Processes designed to improve Bayesian optimization when objective functions exhibit distinct regimes,…

Global: New Transformer Model EVEREST Improves Rare-Event Forecasting AccuracyResearchers have introduced EVEREST, a transformer‑based architecture designed to forecast rare events in multivariate time‑series data. The model was described in a…

Global: New Framework Enhances Noise‑Resilient Retrieval for Edge‑Based LLM AssistantsOverview of the TONEL InitiativeResearchers from multiple institutions introduced Task‑Oriented Noise‑resilient Embedding Learning (TONEL) in a paper posted to arXiv on…

Global: New Two-Stage Framework Boosts Offline Reinforcement Learning in Complex Action SpacesResearchers Matthew Landers, Taylor W. Killian, Thomas Hartvigsen, and Afsaneh Doryab released a preprint on January 7, 2026 that…