Despite thousands of researchers, engineers, and artists actively working on improving text-to-image generation models, systems often fail to produce images that accurately align with the text inputs. We introduce TIFA (Text-to-Image Faithfulness evaluation with question Answering), an automatic evaluation metric that measures the faithfulness of a generated image to its text input via visual question answering (VQA). Specifically, given a text input, we automatically generate several question-answer pairs using a language model. We calculate image faithfulness by checking whether existing VQA models can answer these questions using the generated image. TIFA is a reference-free metric that allows for fine-grained and interpretable evaluations of generated images. TIFA also has better correlations with human judgments than existing metrics. Based on this approach, we introduce TIFA v1.0, a benchmark consisting of 4K diverse text inputs and 25K questions across 12 categories (object, counting, etc.). We present a comprehensive evaluation of existing text-to-image models using TIFA v1.0 and highlight the limitations and challenges of current models. For instance, we find that current text-to-image models, despite doing well on color and material, still struggle in counting, spatial relations, and composing multiple objects. We hope our benchmark will help carefully measure the research progress in text-to-image synthesis and provide valuable insights for further research.

Single polypyrrole (PPy) nanowire-based microfluidic aptasensors were fabricated using a one-step electrochemical deposition method. The successful incorporation of the aptamers into the PPy nanowire was confirmed by fluorescence microscopy image. The microfluidic aptasensor showed responses to IgE protein solutions in the range from 0.01 nM to 100 nM, and demonstrated excellent specificity and sensitivity with faster response and rapid stabilization times (~20 s). At the lowest examined IgE concentration of 0.01nM, the microfluidic aptasensor still exhibited ~0.32% change in the conductance. The functionality of this aptasensor was able to be regenerated using an acid treatment with no major change in sensitivity. In addition, the detection of cancer biomarker MUC1 was performed using another microfluidic aptasensor, which showed a very low detection limit of 2.66 nM MUC1 compared to commercially available MUC1 diagnosis assay (800 nM).

A single nanowire array on a chip with different materials of Palladium, Polypyrrole and Zinc Oxide has been fabricated using electrochemical deposition method. The fabricated single nanowire array has been demonstrated for highly sensitive and specific diagnosis of breast cancer by detecting four volatile organic compound biomarkers: Heptanal, Acetophenone, Isopropyl Myristate and 2-Propanol. The demonstrated sensing limits for Heptanal, Acetophenone, Isopropyl Myristate and 2-propanol using individual Palladium, Polypyrrole and Zinc Oxide nanowires were 8.982 ppm, 798 ppb, 134 ppm, and 129.5 ppm, respectively, and the corresponding sensitivities of resistance change were in the range of 0.3%-5% which indicated excellent sensing performance of the single nanowires. The response time for Palladium, Polypyrrole and Zinc Oxide nanowires to achieve maximum conductance change was less than 200 seconds while also illustrating excellent signal repeatability. With the principal component analysis of the resistance change versus time in each detection period of the nanowire array, the smell prints for the four volatile organic compounds biomarkers of Breast Cancer are discriminated in the 3-D plots.

Bubble‐like domains, typically a precursor to the electrical skyrmions, arise in ultrathin complex oxide ferroelectric–dielectric–ferroelectric heterostructures epitaxially clamped with flat substrates. Here, it is reported that these specially ordered electric dipoles can also be retained in a freestanding state despite the presence of inhomogeneously distributed structural ripples. By probing local piezo and capacitive responses and using atomistic simulations, this study analyzes these ripples, sheds light on how the bubbles are stabilized in the modified electromechanical energy landscape, and discusses the difference in morphology between bubbles in freestanding and as‐grown states. These results are anticipated to be the starting point of a new paradigm for the exploration of electric skyrmions with arbitrary boundaries and physically flexible topological orders in ferroelectric curvilinear space.

No abstract