KLA (Ireland)

End-to-end Visual Question Answering (VQA) models take both an image and a question as input and directly provide an answer to the question as output. Recently, visual program synthesis has introduced a new approach to VQA tasks, wherein a large language model generates a piece of code based on the input question, which is then executed on the input image to produce the answer. Compared to end-to-end models, the visual program synthesis method offers better explainability and flexibility. While existing visual program synthesis methods focus on generating code with correct logic, in this work, we first show that a logically correct code does not always provide the right answer to the input question. Based on this insight, we propose CobraVPS (Code Template Optimization for Better Question Reasoning Accuracy with Visual Program Synthesis) for VQA tasks. Experiments conducted on three different datasets, namely GQA, VQAv2, and Winoground, show that the proposed CobraVPS outperforms the state-of-the-art baseline by up to 5.4% in accuracy. CobraVPS does not require human annotation or model fine-tuning, and demonstrates stable performance across different Code LLMs.

Barrier layers are used in packaging to prevent the diffusion of copper into the solder material, which could otherwise severely limit performance. Barriers should also have low resistivity, good adhesion to the pad and solder metals, and slow reaction kinetics. Nickel has commonly been used as a barrier, but as bump pitches continue to scale down (below 25µm), it becomes unsuitable for use because a thin layer of Ni film can be entirely consumed (and fairly quickly), leaving only an intermetallic compound behind and limiting its function to block electromigration. Thus, there is large interest in new barrier materials at this size. Iron-containing alloys (e.g. FeNi, FeCo) have recently been studied as a barrier material between the copper and solder bumps in lead-free flip chips. In particular, FeCo, demonstrates up to 10 times slower reaction kinetics than pure Ni, thereby significantly reducing intermetallic compound growth and extending its lifetime considerably. The barrier material’s properties are dependent on the composition of the alloy, making the control of electrodeposition process crucial. In this work, we provide for comprehensive analysis of bath components, demonstrated with multiple analytical techniques including UVVIS spectroscopy, HPLC, and electrochemical methods (e.g. voltammetry). Detection of critical breakdown products is also reported. For example, in iron-containing plating baths, dissolved oxygen can easily oxidize Fe(II) to Fe(III). Accumulation of Fe(III) affects the quality and property of the alloy. An accurate and selective method is developed to measure low level of Fe(III) where the Fe(III):Fe(II) ratio is 1:100 or less.

ABSTRACT This file is a PDF of the presentation slides only. The recorded presentation was also made available here .

Estimating the transferability of publicly available pretrained models to a target task has assumed an important place for transfer learning tasks in recent years. Existing efforts propose metrics that allow a user to choose one model from a pool of pre-trained models without having to fine-tune each model individually and identify one explicitly. With the growth in the number of available pre-trained models and the popularity of model ensembles, it also becomes essential to study the transferability of multiple-source models for a given target task. The few existing efforts study transferability in such multi-source ensemble settings using just the outputs of the classification layer and neglect possible domain or task mismatch. Moreover, they overlook the most important factor while selecting the source models, viz., the cohesiveness factor between them, which can impact the performance and confidence in the prediction of the ensemble. To address these gaps, we propose a novel Optimal tranSport-based suBmOdular tRaNsferability metric (OSBORN) to estimate the transferability of an ensemble of models to a downstream task. OSBORN collectively accounts for image domain difference, task difference, and cohesiveness of models in the ensemble to provide reliable estimates of transferability. We gauge the performance of OSBORN on both image classification and semantic segmentation tasks. Our setup includes 28 source datasets, 11 target datasets, 5 model architectures, and 2 pre-training methods. We benchmark our method against current stateof-the-art metrics MS-LEEP and E-LEEP, and outperform them consistently using the proposed approach.

This study aims to improve the ability of teachers in Wonokarto Public Elementary School 2019/2020 Academic Year in classroom management through the application of active knowledge sharing.The study used a School Action Research (PTS) design which was carried out in 2 (two) cycles, carried out in SD Negeri 1 Wonokarto with the research subjects being all SD Negeri 1 Wonokarto teachers in 2019/2020 Academic Year totaling