Skip to main content
Search

George Washington University

5 related articles

Subscribe to updates
  1. JAMA Cardiology

    Coronary inflammation read from noncontrast CT: a fat attenuation index score sharpens risk in people with zero calcium

    Coronary inflammation could previously be measured only on contrast-enhanced coronary CT angiography, while routine calcium scoring uses low-dose noncontrast CT. A deep-learning pipeline trained and externally validated in a multicenter cohort computes a pericoronary fat attenuation index (FAI) score from noncontrast CT; the score performed comparably to the contrast-based FAI and further separated cardiovascular death and major adverse cardiovascular events among people with a calcium score of zero.

  2. arXiv

    SynCo jointly optimizes synthesizer and reasoner in the same online batch, reaching 78.72% average across eight math reasoning benchmarks

    SynCo formulates agentic data synthesis and task solving as a coupled multi-agent reinforcement learning problem, updating a Synthesizer and a Reasoner from the same online interaction batch: the Reasoner receives rollout-level correctness rewards while the Synthesizer receives a task-level reward gated by task quality, answer reliability, and outcome-grounded teachability, with both updates feeding the next synthesis round. Across eight mathematical reasoning benchmarks, SynCo reaches 78.72% average accuracy versus 75.85% for a frozen-Synthesizer variant and 74.96% for the strongest external baseline, ScaleQuest-Math, with gains arising mainly from problems the initial model failed to solve.

  3. arXiv

    Astrophysics gets its own multi-agent system: top-five literature hit rates reach 79.2%, with multiwavelength data on call

    Astrophysics research must juggle fast-growing literature, multiwavelength observations and complex physical models, and AstroGenesis places these in one multi-agent environment where a Supervisor Agent and Planner/Replanner coordinate literature retrieval, observational data analysis, theoretical modeling and research ideation. The current implementation focuses on blazar research, and its Theoretical Modeling Agent uses pretrained neural-network surrogate models to support broadband and multimessenger modeling through natural language. On single-paper and multi-paper literature benchmarks, at least one relevant publication appeared in the top five results for 76.6% and 79.2% of questions.

  4. arXiv

    VideoGen-Agent uses multitask reinforcement learning to teach a video-generation agent to call retrieval, simulation, and verification tools, lifting its base generator from 56.5 to 75.6 on VABench and to 86.1 with upgraded generation tools

    The work introduces VideoGen-Agent, a multimodal agent trained with supervised fine-tuning followed by multitask agentic reinforcement learning to coordinate augmentation, generation, and verification tools across six video-generation tasks (Procedural Knowledge, Single-Entity Identity, Multi-Entity Identity, Physics Simulation, Compositional Scene, and Multi-Shot), and builds VABench, a held-out benchmark of 600 prompts; on VABench it improves the base T2V generator Seedance 1.0 from 56.5 to 75.6, and further to 86.1 when compatible upgraded generation tools unseen during training are used, with human raters preferring that configuration in 100 pairwise comparisons.

  5. Temporal Analysis of Patient-Centered Sentiment in Clinical Notes for Patients With Mental Health Conditions: Retrospective Cohort Study

    Using 16,447 clinical notes from 6,382 patients with mental health diagnoses in the MIMIC-IV database, this study labeled sentiment from patient, physician, and general perspectives with two large language models (DeepSeek-7B and Mistral-7B) and three lexicon-based tools (ClinSent-lexicon, TextBlob, and VADER), finding substantial directional change in sentiment trajectories among patients with multiple admissions, greater fluctuation in Discharge Instructions than in Brief Hospital Course notes, more balanced patient-perspective sentiment versus predominantly neutral physician and general perspectives, better alignment of LLMs with patient-centered annotations than lexicon-based methods, and significantly more negative discharge-note sentiment trajectories among patients who died within 3

You’re all caught up