Hongxuan Wu

Xinyuan Luo, Xingrui Chen, Xunjian Yin et al.

Humanoid robots have achieved impressive locomotion performance, yet contact-rich and long-horizon manipulation remains a major bottleneck. Manipulation is inherently contact-rich and demands compliant whole-body control for stable interaction, while its diversity and long-horizon nature favor modular, planner-compatible interfaces over joint-space tracking. We propose CEER, a compliant end-effector-root (EE-root) control abstraction for modular humanoid loco-manipulation within a hierarchical planning framework. CEER enables compliance-aware whole-body control in an interpretable task space defined by root motion commands and end-effector pose targets, and supports plug-and-play integration with heterogeneous high-level planners. A teacher-student framework is adopted to distill a general motion-tracking controller into a low-level policy that consumes only EE-root commands. We further construct a hierarchical system that integrates heterogeneous planners and task modules through the EE-root interface, enabling diverse manipulation tasks without retraining the underlying whole-body policy. Experiments in simulation and on hardware demonstrate 3.3 cm end-effector tracking accuracy with substantially reduced jerk compared to baselines, stable contact-rich manipulation under teleoperation, and up to 70% success in simulated single-object loco-manipulation tasks within a room-scale environment. These results indicate that compliant EE-root control provides a practical abstraction for humanoid loco-manipulation, enabling modular and scalable integration of diverse skills.

Jiabao Brad Wang, Siyuan Cao, Hongxuan Wu et al.

Selecting an effective docking algorithm is highly context-dependent, and no single method performs reliably across structural, chemical, or protocol regimes. We introduce MolAS, a lightweight algorithm selection system that predicts per-algorithm performance from pretrained protein-ligand embeddings using attentional pooling and a shallow residual decoder. With only hundreds to a few thousand labelled complexes, MolAS achieves up to 15% absolute improvement over the single-best solver (SBS) and closes 17-66% of the Virtual Best Solver (VBS)-SBS gap across five diverse docking benchmarks. Analyses of reliability, embedding geometry, and solver-selection patterns show that MolAS succeeds when the oracle landscape exhibits low entropy and separable solver behaviour, but collapses under protocol-induced hierarchy shifts. These findings indicate that the main barrier to robust docking AS is not representational capacity but instability in solver rankings across pose-generation regimes, positioning MolAS as both a practical in-domain selector and a diagnostic tool for assessing when AS is feasible.

Hongxuan Wu, Yukun Zhang, Xueqing Zhou

When a multimodal Transformer answers a visual question, is the prediction driven by visual evidence, linguistic reasoning, or genuinely fused cross-modal computation -- and how does this structure evolve across layers? We address this question with a layer-wise framework based on Partial Information Decomposition (PID) that decomposes the predictive information at each Transformer layer into redundant, vision-unique, language-unique, and synergistic components. To make PID tractable for high-dimensional neural representations, we introduce \emph{PID Flow}, a pipeline combining dimensionality reduction, normalizing-flow Gaussianization, and closed-form Gaussian PID estimation. Applying this framework to LLaVA-1.5-7B and LLaVA-1.6-7B across six GQA reasoning tasks, we uncover a consistent \emph{modal transduction} pattern: visual-unique information peaks early and decays with depth, language-unique information surges in late layers to account for roughly 82\% of the final prediction, and cross-modal synergy remains below 2\%. This trajectory is highly stable across model variants (layer-wise correlations $>$0.96) yet strongly task-dependent, with semantic redundancy governing the detailed information fingerprint. To establish causality, we perform targeted Image$\rightarrow$Question attention knockouts and show that disrupting the primary transduction pathway induces predictable increases in trapped visual-unique information, compensatory synergy, and total information cost -- effects that are strongest in vision-dependent tasks and weakest in high-redundancy tasks. Together, these results provide an information-theoretic, causal account of how vision becomes language in multimodal Transformers, and offer quantitative guidance for identifying architectural bottlenecks where modality-specific information is lost.

Edward Kim, Yuri Cho, Jose Eduardo E. Lima et al.

Digital health interventions increasingly deliver home exercise programs via sensor-equipped devices such as smartphones, enabling remote monitoring of adherence and performance. However, current software is usually authored before clinical encounters as libraries of modules for broad impairment categories. At the point of care, clinicians can only choose from these modules and adjust a few parameters (for example, duration or repetitions). As a result, individual limitations, goals, and environmental constraints are often not reflected, limiting personalization and benefit. We propose a paradigm in which large language models (LLMs) act as constrained translators that convert clinicians' exercise prescriptions into intervention software. Clinicians remain the decision makers: they design exercises during the encounter, tailored to each patient's impairments, goals, and environment, and the LLM generates matching software. We conducted a prospective single-arm feasibility study with 20 licensed physical and occupational therapists who created 40 individualized upper extremity programs for a standardized patient; 100% of prescriptions were translated into executable software, compared with 55% under a representative template-based digital health intervention (p < 0.01). LLM-generated software correctly delivered 99.7% of instructions and monitored performance with 88.4% accuracy (95% confidence interval, 0.843-0.915). Overall, 90% of therapists judged the system safe for patient interaction and 75% expressed willingness to adopt it in practice. To our knowledge, this is the first prospective evaluation of clinician-directed intervention software generation with an LLM in health care, demonstrating feasibility and motivating larger trials in real patient populations.

Siyuan Cao, Hongxuan Wu, Jiabao Brad Wang et al.

Molecular docking is a core tool in drug discovery for predicting ligand-target interactions. Despite the availability of diverse search-based and machine learning approaches, no single docking algorithm consistently dominates, as performance varies by context. To overcome this challenge, algorithm selection frameworks such as GNNAS-Dock, built on graph neural networks, have been proposed. This study introduces an enhanced system, MC-GNNAS-Dock, with three key advances. First, a multi-criteria evaluation integrates binding-pose accuracy (RMSD) with validity checks from PoseBusters, offering a more rigorous assessment. Second, architectural refinements by inclusion of residual connections strengthen predictive robustness. Third, rank-aware loss functions are incorporated to sharpen rank learning. Extensive experiments are performed on a curated dataset containing approximately 3200 protein-ligand complexes from PDBBind. MC-GNNAS-Dock demonstrates consistently superior performance, achieving up to 5.4% (3.4%) gains under composite criteria of RMSD below 1Å (2Å) with PoseBuster-validity compared to the single best solver (SBS) Uni-Mol Docking V2.