Vision-language models (VLMs), such as CLIP, have gained significant popularity as foundation models, with numerous fine-tuning methods developed to enhance performance on downstream tasks. However, due to their inherent vulnerability and the common practice of selecting from a limited set of open-source models, VLMs suffer from a higher risk of adversarial attacks than traditional vision models. Existing defense techniques typically rely on adversarial fine-tuning during training, which requires labeled data and lacks of flexibility for downstream tasks. To address these limitations, we propose robust test-time prompt tuning (R-TPT), which mitigates the impact of adversarial attacks during the inference stage. We first reformulate the classic marginal entropy objective by eliminating the term that introduces conflicts under adversarial conditions, retaining only the pointwise entropy minimization. Furthermore, we introduce a plug-and-play reliability-based weighted ensembling strategy, which aggregates useful information from reliable augmented views to strengthen the defense. R-TPT enhances defense against adversarial attacks without requiring labeled training data while offering high flexibility for inference tasks. Extensive experiments on widely used benchmarks with various attacks demonstrate the effectiveness of R-TPT. The code is available in https://github.com/TomSheng21/R-TPT.
Property measurements of individual nanowires are challenging due to the small sizes of the structures. Scanning probe microscopy has thus far been the dominant tool for the characterization of nanomaterial properties. We have developed an alternative novel approach that allows a direct measurement of the mechanical properties of individual nanowires by in situ transmission electron microscopy (TEM). The technique relies on the electric field induced mechanical resonance of the nanowire observed in TEM, as it directly correlates the atomic‐scale microstructure of the nanowire with its physical properties. In this paper, the measurement of Young's modulus for composite SiO2/SiC wires is reported. The experimental results are in agreement with theoretically expected values.
In recent years, great advances in pre-trained language models (PLMs) have sparked considerable research focus and achieved promising performance on the approach of dense passage retrieval, which aims at retrieving relative passages from massive corpus with given questions. However, most of existing datasets mainly benchmark the models with factoid queries of general commonsense, while specialised fields such as finance and economics remain unexplored due to the deficiency of large-scale and high-quality datasets with expert annotations. In this work, we propose a new task, policy retrieval, by introducing the Chinese Stock Policy Retrieval Dataset (CSPRD), which provides 700+ prospectus passages labeled by experienced experts with relevant articles from 10k+ entries in our collected Chinese policy corpus. Experiments on lexical, embedding and fine-tuned bi-encoder models show the effectiveness of our proposed CSPRD yet also suggests ample potential for improvement. Our best performing baseline achieves 56.1% MRR@10, 28.5% NDCG@10, 37.5% Recall@10 and 80.6% Precision@10 on dev set.
Additional file 1: Table S1 High expressed pathways in the result of Gene set enrichment analysis (GSEA). Table S2 Low expressed pathways in the result of Gene set enrichment analysis (GSEA). Table S3 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis and Gene Ontology (GO) annotations of TIMM8A associated proteins. Table S4 The potential upstream miRNAs of TIMM8A. Table S5 The candidate circRNAs for hsa-miR-34c-5p. Table S6 The candidate circRNAs for hsa-miR-449a.
