We present a new problem: grounding natural language instructions to mobile user interface actions, and create three new datasets for it. For full task evaluation, we create PIX-ELHELP, a corpus that pairs English instructions with actions performed by people on a mobile UI emulator. To scale training, we decouple the language and action data by (a) annotating action phrase spans in HowTo instructions and (b) synthesizing grounded descriptions of actions for mobile user interfaces. We use a Transformer to extract action phrase tuples from long-range natural language instructions. A grounding Transformer then contextually represents UI objects using both their content and screen position and connects them to object descriptions. Given a starting screen and instruction, our model achieves 70.59% accuracy on predicting complete ground-truth action sequences in PIXELHELP.
Efficient recovery of dyes from textile
wastewaters using membrane-based
technologies calls for membranes with low rejection of salts but high
rejection of dyes. In this study, we intercalated the rigid 2D-ultrathin
MXene (Ti3C2T
x
)
nanosheet into the loose framework assembled by 1D aramid nanofibers
and obtained a MXene-reinforced aramid nanofiber (ANF) membrane. The
optimized MXene-reinforced ANF membrane (20 wt % MXene loading) exhibited
a high selectivity of dye/salt, that is, 99.1 ± 0.3% for Alcian
Blue, 96.1 ± 2.4% for Congo Red, 98.6 ± 0.9% for Rose Bengal,
and extremely low rejection of salts (<0.2%) at a concentration
of 5 g L–1 of NaCl with separation factors of 826,
1068, and 758 for Alcian Blue, Congo Red, and Rose Bengal, respectively.
Moreover, the membrane permeability could reach 195.3 ± 6.5 L
m–2 h–1 bar–1. The MXene-reinforced ANF membrane demonstrated a superior separation
performance to most of the other reported membranes for dye/salt fraction.
Mechanistic investigation showed that the reinforced rigidity of the
membrane by the intercalated MXene was the main mechanism for the
enhanced permeability and selectivity of the MXene/ANF composite membrane
(that is, a nanochannel of 2.34 nm was maintained for the MXene/ANF
composite membrane after filtration while the nanochannel of the ANF
membrane was increased from 2.27 to 2.43 nm). The results highlight
that the nanofibrous membranes reinforced by rigid 2D materials are
promising for an effective separation of dyes and salts in textile
wastewaters.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.