Search for: All records

Fine-grained entity typing (FET) is the task of identifying specific entity types at a fine-grained level for entity mentions based on their contextual information. Conventional methods for FET require extensive human annotation, which is time-consuming and costly given the massive scale of data. Recent studies have been developing weakly supervised or zero-shot approaches.We study the setting of zero-shot FET where only an ontology is provided. However, most existing ontology structures lack rich supporting information and even contain ambiguous relations, making them ineffective in guiding FET. Recently developed language models, though promising in various few-shot and zero-shot NLP tasks, may face challenges in zero-shot FET due to their lack of interaction with task-specific ontology. In this study, we propose OnEFET, where we (1) enrich each node in the ontology structure with two categories of extra information: instance information for training sample augmentation and topic information to relate types with contexts, and (2) develop a coarse-to-fine typing algorithm that exploits the enriched information by training an entailment model with contrasting topics and instance-based augmented training samples. Our experiments show that OnEFET achieves high-quality fine-grained entity typing without human annotation, outperforming existing zero-shot methods by a large margin and rivaling supervised methods. OnEFET also enjoys strong transferability to unseen and finer-grained types. Code is available at https://github.com/ozyyshr/OnEFET. 

For decades, infrared (IR) spectroscopy has advanced on two distinct frontiers: enhancing spatial resolution and broadening spectroscopic information. Although atomic force microscopy (AFM)-based IR microscopy overcomes Abbe’s diffraction limit and reaches sub-10 nm spatial resolutions, time-domain two-dimensional IR spectroscopy (2DIR) provides insights into molecular structures, mode coupling and energy transfers. Here we bridge the boundary between these two techniques and develop AFM-2DIR nanospectroscopy. Our method offers the spatial precision of AFM in combination with the rich spectroscopic information provided by 2DIR. This approach mechanically detects the sample’s photothermal responses to a tip-enhanced femtosecond IR pulse sequence and extracts spatially resolved spectroscopic information via FFTs. In a proof-of-principle experiment, we elucidate the anharmonicity of a carbonyl vibrational mode. Further, leveraging the near-field photons’ high momenta from the tip enhancement for phase matching, we photothermally probe hyperbolic phonon polaritons in isotope-enriched h10BN. Our measurements unveil an energy transfer between phonon polaritons and phonons, as well as among different polariton modes, possibly aided by scattering at interfaces. The AFM-2DIR nanospectroscopy enables the in situ investigations of vibrational anharmonicity, coupling and energy transfers in heterogeneous materials and nanostructures, especially suitable for unravelling the relaxation process in two-dimensional materials at IR frequencies. 

Human expectations arise from their understanding of others and the world. In the context of human-AI interaction, this understanding may not align with reality, leading to the AI agent failing to meet expectations and compromising team performance. Explicable planning, introduced as a method to bridge this gap, aims to reconcile human expectations with the agent's optimal behavior, facilitating interpretable decision-making. However, an unresolved critical issue is ensuring safety in explicable planning, as it could result in explicable behaviors that are unsafe. To address this, we propose Safe Explicable Planning (SEP), which extends the prior work to support the specification of a safety bound. The goal of SEP is to find behaviors that align with human expectations while adhering to the specified safety criterion. Our approach generalizes the consideration of multiple objectives stemming from multiple models rather than a single model, yielding a Pareto set of safe explicable policies. We present both an exact method, guaranteeing finding the Pareto set, and a more efficient greedy method that finds one of the policies in the Pareto set. Additionally, we offer approximate solutions based on state aggregation to improve scalability. We provide formal proofs that validate the desired theoretical properties of these methods. Evaluation through simulations and physical robot experiments confirms the effectiveness of our approach for safe explicable planning. 

As a treatment for the widely spread cardiovascular diseases (CVD), bypass vascular grafts have room for improvement in terms of mechanical property match with native arteries. A 3D‐printed nozzle is presented, featuring unique internal structures, to extrude artificial vascular grafts with a flower‐mimicking geometry. The multilayer‐structured graft wall allows the inner and outer layers to interfere sequentially during lateral expansion, replicating the nonlinear elasticity of native vessels. Both experiment and simulation results verify the necessity and benefit of the flower‐mimicking structure in obtaining the self‐toughening behavior. The gelation study of natural polymers and the utilization of sacrificial phase enables the smooth extrusion of the multiphase conduit, where computer‐assisted image analysis is employed to quantify manufacturing fidelity. The cell viability tests demonstrate the cytocompatibility of the gelatin methacryloyl (GelMA)/sodium alginate grafts, suggesting potential for further clinical research with further developments. This study presents a feasible approach for fabricating bypass vascular grafts and inspires future treatments for CVD.