Brigham and Women's Hospital · Harvard Medical School

XLab

Micro-to-Macro Mechanistic AI

We build AI that links micro-scale mechanisms—model parameters, atomic interactions, molecular signals—to macro-scale behavior, across fundamental AI, life science, and materials science.

Led by Dr. Hao Xu, Investigator at Brigham and Women's Hospital and Instructor in Medicine at Harvard Medical School.

Research Publications Team
Recent work in ACL 2026ECCV 2026Genome BiologyJ. Cheminform.Communications Chemistry

News

Jul 31, 2026 New paper published in Journal of Cheminformatics: How Well Can Off-the-Shelf LLMs Elucidate Molecular Structures from Mass Spectra Using Chain-of-Thought Reasoning?
Jun 24, 2026 New preprint: Staying VIGILant: Mitigating Visual Laziness via Counterfactual Visual Alignment in MLLMs. Accepted to ECCV 2026.
Jun 09, 2026 New preprint on bioRxiv: ExplainBind: Explainable Physicochemical Determinants of Protein-Ligand Binding via Non-Covalent Interactions
May 02, 2026 New paper published in Genome Biology: EvoRMD: Integrating Biological Context and Evolutionary RNA Language Models for Interpretable Prediction of RNA Modifications.
Apr 04, 2026 New preprint: Not All Directions Matter: Toward Structured and Task-Aware Low-Rank Adaptation, accepted to ACL 2026.
Oct 03, 2025 New preprint: HyperAdaLoRA: Accelerating LoRA Rank Allocation During Training via Hypernetworks without Sacrificing Performance, accepted to ACL Findings 2026.

Research Tracks

Research →
TRACK IFundamental AI
From parameter dynamics to reliable model behavior
  1. 1Parameter DynamicsParameter sensitivity and structure; efficient, task-aware adaptation
  2. 2Multimodal UnderstandingVisual grounding and cross-modal reasoning
  3. 3Visual AlignmentCounterfactual alignment; mitigating visual laziness and modality bias
  4. 4Reliable AI SystemsRobust, efficient multimodal intelligence across tasks and domains
TRACK IILife Science
From molecular mechanisms to health outcomes
  1. 1Molecular MechanismsPhysicochemical drivers of binding, catalysis, and conformational change
  2. 2Cellular SystemsMolecular networks, pathways, regulation, and cell states
  3. 3Organismal MechanismsIntegrated physiology, disease, and treatment response
  4. 4Health OutcomesClinical and population-level diagnosis, prognosis, and intervention
TRACK IIIMaterials Science
From atomic interactions to materials discovery and impact
  1. 1Atomic InteractionsAtomic structure, bonding, electrons, and defects
  2. 2Structure–Property MechanismsSurfaces, phases, transport, and stability
  3. 3Materials Discovery & CatalysisML-potential–guided search and generative design
  4. 4Applications & ImpactMaterials for energy, environment, electronics, and beyond
OUR MISSION Understand mechanisms Shape behaviors Enable discovery

Selected Publications

All publications →
  1. bioRxiv
    ExplainBind: Explainable Physicochemical Determinants of Protein-Ligand Binding via Non-Covalent Interactions
    Z. Meng†, Z. Bai†, K. Yuan, J. Song, Y. Zhang, J. H. Cheah, W. Jiang, A. Skepner, K. J. Leahy, I. Ounis, W. M. Oldham*, Z. Meng*, Hao Xu*, and J. Loscalzo
    bioRxiv, 2026
    HTML
  2. Genome Biol.
    EvoRMD: Integrating Biological Context and Evolutionary RNA Language Models for Interpretable Prediction of RNA Modifications
    B. Wang, H. Zhang, T. Cui, X. Wang, J. Song*, and Hao Xu*
    Genome Biology, 2026
    HTML
  3. ACL
    Not All Directions Matter: Toward Structured and Task-Aware Low-Rank Adaptation
    X. Xiao, C. Ma, Y. Zhang, C. Liu, Z. Wang, Y. Li, L. Zhao, G. Hu, T. Wang, and Hao Xu
    In Association for Computational Linguistics (ACL), 2026
    HTML
  4. JCIM
    Advancing Drug Discovery with Enhanced Chemical Understanding via Asymmetric Contrastive Multimodal Learning
    Yifei Wang†, Yunrui Li†, Lin Liu, Pengyu Hong, and Hao Xu*
    Journal of Chemical Information and Modeling, 2025
    HTML
  5. Chem
    Enantioselective Radical N-Heterobicyclization by New Mode of Asymmetric Induction via Kinetically Stable Chiral Radical Center
    Hao Xu, Duo-Sheng Wang, Zhen-Yu Zhu, Arghya Deb, and X. Peter Zhang*
    Chem, 2024
    HTML

Work with XLab

We welcome collaborations with students and researchers across AI, chemistry, and medicine. Reach out at haoxu0303@gmail.com.