Fant 10273 publikasjoner. Viser side 176 av 411:
2002
There is considerable interest in identifying chemicals which have the potential to undergo long-range environmental transport (LRTP), accumulate in remote regions, and represent a possible risk to environmental and human health. In this report, we have screened a list of 1,000 organic chemicals, as well as selected brominated dioxins and furans (PBDD/Fs), for their potential to be dispersed, transferred to, and accumulated in remote regions. This screening was carried out applying a new set of LRTP metrics, collectively referred to as the emissions fractions approach (EFA), as implemented in a modified version of the OECD POV and LRTP (long-range transport potential) Screening Tool (The Tool).
NILU
2023
2006
2002
2023
2015
2011
Identifying trends in the ocean wave climate by time series analyses of significant wave heightdata.
2013
2020
Ikke-spesifikk screening av støv fra norske husholdninger
Denne rapporten presenterer resultater fra en ikke-spesifikk screening av husstøv fra norske hjem. Totalt ble 203 kjemiske forbindelser identifisert, med ftalater som den mest dominerende stoffgruppen. Flere av de påviste stoffene er kjent som hormonforstyrrende, nevrotoksiske eller klassifisert som persistente, mobile og toksiske (PMT). Resultatene viser et endret stoffmønster sammenlignet med tidligere studier og understreker behovet for videre overvåkning av innemiljø, forskning på cocktail-effekter og bedre regulering av forbrukerprodukter.
NILU
2025
2005
Cross-modal representation learning aims to learn a shared representation space where data from multiple modalities can be effectively compared, fused, and understood. This paper investigates the role of increased diversity in the similarity score matrix in enhancing the performance of the CLIP (Contrastive Language-Image Pretraining), a multi-modal learning model that establishes a connection between images and text within a joint embedding space. Two transforming approaches, sine and sigmoid (including two versions), are incorporated into the CLIP model to amplify larger values and diminish smaller values within the similarity matrix (logits). Hardware limitations are addressed using a more compact text encoder (DistilBERT) and a pre-trained ResNet50 image encoder. The proposed adaptations are evaluated on various benchmarks, including image classification and image/text retrieval tasks, using 10 benchmark datasets such as Food101, Flickr30k, and COCO. The performance of the adapted models is compared to the base CLIP model using Accuracy, mean per class, and Recall@k metrics. The results demonstrate improvements in Accuracy (up to 5.32% enhancement for the PatchCamelyon dataset), mean per class (up to 14.48% enhancement for the FGVCAircraft dataset), and retrieval precision (with an increase of up to 45.20% in Recall@1 for the COCO dataset), compared to the baseline algorithm (CLIP).
2023
2012
2008
2015
2017
Impact assessment for emissions to air from a planned aluminium smelter i Reydarfjordur, Iceland. NILU OR
2001
2004