Search for: All records

Phononic waveguides (PnWGs) are devices with rationally designed periodic structures to manipulate mechanical oscillations and to engineer and control the propagation of acoustic waves, thus allowing for frequency and band selection of wave transmission and routing, promising for both classical and quantum transduction on chip-scale platforms with various constituent materials of interest. They can be incorporated into both electromechanical and optomechanical signal transduction schemes. Here, we present an overview of emerging micro/nanoscale PnWGs and offer perspectives for future. We evaluate the typical structural designs, frequency scaling, and phononic band structures of the PnWGs. Material choices, fabrication techniques, and characterization schemes are discussed based on different PnWG designs. For classical transduction schemes, an all-phononic integrated circuit perspective is proposed. Toward emerging quantum applications, the potential of utilizing PnWGs as universal interfaces and transduction channels has been examined. We envision PnWGs with extraordinary propagation properties, such as nonreciprocity and active tunability, can be realized with unconventional design strategies (e.g., inverse design) and advanced materials (e.g., van der Waals layered crystals), opening opportunities in both classical and quantum signal transduction schemes.

 

Visual question answering (VQA) requires systems to perform concept-level reasoning by unifying unstructured (e.g., the context in question and answer; “QA context”) and structured (e.g., knowledge graph for the QA context and scene; “concept graph”) multimodal knowledge. Existing works typically combine a scene graph and a concept graph of the scene by connecting corresponding visual nodes and concept nodes, then incorporate the QA context representation to perform question answering. However, these methods only perform a unidirectional fusion from unstructured knowledge to structured knowledge, limiting their potential to capture joint reasoning over the heterogeneous modalities of knowledge. To perform more expressive reasoning, we propose VQA-GNN, a new VQA model that performs bidirectional fusion between unstructured and structured multimodal knowledge to obtain unified knowledge representations. Specifically, we inter-connect the scene graph and the concept graph through a super node that represents the QA context, and introduce a new multimodal GNN technique to perform inter-modal message passing for reasoning that mitigates representational gaps between modalities. On two challenging VQA tasks (VCR and GQA), our method outperforms strong baseline VQA methods by 3.2% on VCR (Q-AR) and 4.6% on GQA, suggesting its strength in performing concept-level reasoning. Ablation studies further demonstrate the efficacy of the bidirectional fusion and multimodal GNN method in unifying unstructured and structured multimodal knowledge. 

This work reports experimental demonstrations of reversible crystalline phase transition in ultrathin molybdenum ditelluride (MoTe₂) controlled by thermal and mechanical mechanisms on the van der Waals (vdW) nanoelectromechanical systems (NEMS) platform, with hexagonal boron nitride encapsulated MoTe₂structure residing on top of graphene layer. Benefiting from very efficient electrothermal heating and straining effects in the suspended vdW heterostructures, MoTe₂phase transition is triggered by rising temperature and strain level. Raman spectroscopy monitors the MoTe₂crystalline phase signatures in situ and clearly records reversible phase transitions between hexagonal 2H (semiconducting) and monoclinic 1T′ (metallic) phases. Combined with Raman thermometry, precisely measured nanomechanical resonances of the vdW devices enable the determination and monitoring of the strain variations as temperature is being regulated by electrothermal control. These results not only deepen the understanding of MoTe₂phase transition, but also demonstrate a novel platform for engineering MoTe₂phase transition and multiphysical devices.