Skip to main content
Search

Yale University

26 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

    Brain-metastasis segmentation adapts to new hospitals, modalities and lesion classes: FGCARL reaches 68.28% mean Dice on BraTS-METS 2023

    Clinical images arrive in batches by institution, modality and annotated structure, so a segmentation model must absorb a new imaging distribution and new lesion classes without forgetting what it already learned. FGCARL puts bidirectional feature-statistics alignment and Fisher-weighted parameter constraints into one incremental framework that needs no replay of historical patient data. Moving from UCSF-BMSR to BraTS-METS 2023, where domain and classes shift together, FGCARL with Swin UNETR reaches 68.28% mean Dice across the three tumor regions of the new domain, against 60.94% for the strongest comparator, CLAMTS.

  3. bioRxiv

    SVC treats each gene's intracellular spread as an image patch: other genes in the same cell supply the context for predicting where unmeasured transcripts sit

    Subcellular-resolution spatial transcriptomics can locate transcripts inside cells, but earlier methods targeted single analysis tasks and lacked a unified framework for characterizing and predicting gene localization in a cell-specific, context-aware way. SVC represents each gene's within-cell spatial expression as an image patch and uses other genes in the same cell as context, learning cross-gene spatial dependencies through self-supervised masking. On seqFISH+ mouse embryonic fibroblast data, SVC beat four image-based baselines on all six metrics for predicting subcellular patterns of unmeasured genes, with median gains of 5.2%–21.0% over the more challenging baseline pair.

  4. bioRxiv

    Paired transcriptome and 3D genome from one cell: scHiR fuses two sparse modalities into a single cell embedding

    Technologies that simultaneously sequence single-cell transcriptome and 3D genome already yield paired measurements from the same cell, but the two modalities differ sharply in noise, scale and biological meaning, and were previously analyzed separately with latent spaces that need not align. scHiR maps gene expression and chromatin contact matrices into one hyperspherical latent space through modality-specific encoders and reconstructs both signals. On paired data from HiRES, GAGE-seq and LiMCA, the joint embedding matched or exceeded single-omics methods in clustering, and on mouse main olfactory epithelium it raised the Spearman correlation between pseudotime and developmental stage from 0.357 and 0.205 for single-omics embeddings to 0.517.

  5. Reward the search, not just the answer: process supervision lifts retrieval recall and holds up out of domain

    Search agents for multi-hop information seeking are trained on step-level judgments of each search action instead of only on final-answer correctness. RAG-Gym casts search as a Markov decision process and compares architecture, parameter optimization and action evaluation, combining a reasoning-reflection architecture, preference optimization and critic-guided inference into Re2Search++. Trained on HotpotQA, it beats outcome-supervised baselines on the unseen Bamboogle benchmark with higher retrieval recall at comparable or fewer search steps, and a critic trained on Llama-3.1-8B raises GPT-4o-mini MedQA accuracy.

  6. arXiv

    Same attention heads, new task, new role: one internal activation edit changes which fact the model reads and how it influences later layers

    Across five open-weight models, holding graph facts fixed and changing only the task instruction redistributes how the same attention heads influence later computation; in Gemma-2-9B and Qwen3-14B, editing one early internal activation vector switches which fact the model reads while the written instruction and weights stay fixed.

  7. arXiv

    Agents learn online without only reading forward: revising retrieved reflections from trajectories beats memory and prompt optimization on three benchmarks

    Agents running over long interaction streams must adapt online to distribution shifts, yet existing agent harnesses either evolve their state through forward information flow alone or rely on extra replay or validation data for optimization. ReHarness combines forward read and write with backward revision under a strictly online streaming protocol: after each instance, the resulting trajectory becomes feedback that revises the reflections retrieved for that instance, propagating feedback backward through the textual harness state. Across three continual learning benchmarks and three language model backbones, it consistently outperforms raw-context baselines, memory- and reflection-based methods, and prompt optimization approaches.

  8. arXiv

    SignRAG combines hierarchical pretraining with retrieval-augmented reinforcement fine-tuning to make a gloss-free SLT model beat gloss-supervised methods on all CSL-Daily metrics

    SignRAG is a unified gloss-free sign language translation framework that first learns linguistically grounded sign representations via fine-grained sign-text alignment, then jointly pretrains the sign encoder with Qwen3-8B, adds a target-domain-only retrieval gallery for instance-specific translation cues, and applies RUG-RFT combining translation-quality and retrieval-utility rewards to suppress harmful reliance; it sets new state-of-the-art results on CSL-Daily, PHOENIX-2014T, How2Sign, and OpenASL, and is reported as the first gloss-free approach to surpass gloss-supervised methods across all reported metrics on CSL-Daily.

  9. arXiv

    Researchers resummed the back-to-back energy-energy correlator plateau from weak to strong coupling and completed the N3LL calculation with an agentic workflow

    The work resums the leading- and next-to-leading-power logarithms of the back-to-back energy-energy correlator in N=4 supersymmetric Yang-Mills theory through N3LL accuracy, including non-planar contributions, obtains a new next-to-leading-power function, and combines it with conformal bootstrap and strong-coupling supergravity data to construct an asymptotically constrained interpolation of the plateau height from weak to strong coupling; the calculation was implemented with a human-in-the-loop agentic workflow.

  10. arXiv

    SoloQ quantizes diffusion language models to 4 bits without calibration, holding accuracy on LLaDA, Dream, Fast-dLLM v2 and Nemotron-Labs-Diffusion while beating calibration-based baselines

    The authors present SoloQ, which uses a K-RPBH rotation to map weights and activations into a predictable near-Gaussian basis plus a scale-only bias correction for calibration-free 4-bit quantization, and quantizes the KV cache at block commit for block-diffusion models; across LLaDA-Base, LLaDA-1.5, Dream-7B, Fast-dLLM v2 and Nemotron-Labs-Diffusion, SoloQ-C and SoloQ-N retain accuracy under W4A4 and W4A4KV4 and exceed calibration-based baselines such as DLLMQuant++, STaR-Quant and FAIR-Calib on average, with the NVFP4 path also cutting peak memory and speeding up end-to-end inference.