Machine Learning Science and Technology,
Journal Year:
2024,
Volume and Issue:
5(3), P. 035076 - 035076
Published: Sept. 1, 2024
Abstract
Gravitational
lensing
data
is
frequently
collected
at
low
resolution
due
to
instrumental
limitations
and
observing
conditions.
Machine
learning-based
super-resolution
techniques
offer
a
method
enhance
the
of
these
images,
enabling
more
precise
measurements
effects
better
understanding
matter
distribution
in
system.
This
enhancement
can
significantly
improve
our
knowledge
mass
within
galaxy
its
environment,
as
well
properties
background
source
being
lensed.
Traditional
typically
learn
mapping
function
from
lower-resolution
higher-resolution
samples.
However,
methods
are
often
constrained
by
their
dependence
on
optimizing
fixed
distance
function,
which
result
loss
intricate
details
crucial
for
astrophysical
analysis.
In
this
work,
we
introduce
DiffLense
,
novel
pipeline
based
conditional
diffusion
model
specifically
designed
gravitational
images
obtained
Hyper
Suprime-Cam
Subaru
Strategic
Program
(HSC-SSP).
Our
approach
adopts
generative
model,
leveraging
detailed
structural
information
present
Hubble
space
telescope
(HST)
counterparts.
The
trained
generate
HST
data,
conditioned
HSC
pre-processed
with
denoising
thresholding
reduce
noise
interference.
process
leads
distinct
less
overlapping
during
model’s
training
phase.
We
demonstrate
that
outperforms
existing
state-of-the-art
single-image
techniques,
particularly
retaining
fine
necessary
analyses.
Abstract
The
thermal
phase
transition
characteristics
of
vanadium
dioxide
can
effectively
regulate
the
absorption
performance
metamaterial
absorbers.
A
petal‐shaped
absorber
is
designed
based
on
control
and
prepared
samples
using
surface
projection
micro
stereolithography
(PµSL)
3D
printing
technology.
Then,
thin
films
are
coated
ultrasonic
spraying
method.
PµSL
technology
makes
it
possible
to
prepare
high‐precision
complex
samples,
method
enables
coating
at
room
temperature.
To
demonstrate
stability
this
method,
two
different
structures
prepared:
planar
structure
hemispherical
structure.
In
experiment,
a
non‐contact
heating
used
temperature
film,
which
made
sample
controllable,
stable,
uniformly
heated.
experimental
results
show
that
has
excellent
modulation
switching
characteristics,
proves
reliability
Journal of Applied Physics,
Journal Year:
2025,
Volume and Issue:
137(5)
Published: Feb. 3, 2025
The
metalens
discussed
in
this
article
is
a
planar
optical
device
built
on
metasurfaces,
utilizing
precisely
engineered
subwavelength
nanostructured
wires
to
manipulate
electromagnetic
waves.
This
enables
precise
control
over
the
phase,
amplitude,
and
polarization
of
these
Recent
advancements
research
primarily
focus
their
design
principles,
performance
optimization,
imaging
enhancement,
manufacturing
techniques,
various
application
domains.
reviews
progress
development,
highlighting
principles
across
different
frequency
bands
applications
science
technology.
Future
directions
are
expected
emphasize
material
innovation,
simplification
processes,
further
improvements
drive
adoption
cutting-edge
fields.
JETP Letters,
Journal Year:
2025,
Volume and Issue:
unknown
Published: March 24, 2025
A
transmission-type
all-dielectric
coded
supergrating
operating
in
the
terahertz
band
is
designed.
The
structure
uses
Si
as
dielectric
material
to
realize
change
of
transmission
phase
2π
0.06
~
0.15
THz.
Compared
with
conventional
gratings,
supergratings
have
advantages
wide
bandwidth,
high
transmittance,
easy
coupling
fiber,
and
low
cost.
It
represents
a
new
direction
optical
coding
technology
has
application
prospect
research
value.
We
design
1-bit
supergrating.
By
changing
width
grating
element
structure,
two
basic
structures
coefficients
continuous
changes
180°
are
obtained.
Based
on
addition
theorem
digital
coding,
far-field
scattering
angle
controlled
by
encoding
unit
grating,
its
arrangement
order
physical
superposition.
generalized
Snell
principle,
designed
calculated
theoretically,
characteristics
simulation
results
analyzed.
Nano Letters,
Journal Year:
2025,
Volume and Issue:
unknown
Published: March 31, 2025
Enantioselective
sorting
at
the
nanoscale
is
highly
significant
in
fields
such
as
medical
research,
material
science,
and
drug
development.
However,
previous
studies
mainly
focus
on
static
chiral
particle
separation,
hindering
practical
applications.
Here,
we
utilize
synergy
between
enantioselective
potential
wells
flow
to
sort
nanoparticle
enantiomers
a
high
velocity
of
800
μm/s.
An
enhanced
field
induced
by
mirror-enhanced
toroidal
dipole
employed
amplify
gradient
force,
creating
separated
for
opposite
chirality.
By
regulating
force
fluidic
drag
achieve
20
200
nm
particles
with
separation
distances
larger
than
32
48
μm
area
70.96
×
μm2,
respectively.
Furthermore,
efficient
also
demonstrated
their
wells.
Our
work
holds
tremendous
biotechnology,
nanotechnology,
pharmacology.
ChemPhotoChem,
Journal Year:
2025,
Volume and Issue:
unknown
Published: May 1, 2025
Chirality
plays
a
fundamental
role
in
molecular
sciences,
with
chiroptical
properties
offering
valuable
insights
into
the
interaction
between
chiral
molecules
and
polarized
light.
Designing
materials
enhanced
requires
deep
understanding
of
underlying
physical
principles,
often
revealed
only
through
large
datasets.
In
this
context,
artificial
intelligence
(AI)
emerges
as
powerful
tool
for
accelerating
discovery
optimization,
efficiently
exploring
vast
chemical
spaces.
This
work
explores
synergy
AI
properties,
highlighting
recent
advances
data‐driven
approaches
circular
dichroism
circularly
luminescence.
has
demonstrated
its
ability
to
predict
these
phenomena
accurately
while
uncovering
structure–property
relationships
that
can
remain
hidden
under
traditional
methods.
Various
strategies
are
examined
integrating
challenges
future
directions
field
discussed.
conclusion,
combining
intuition
offers
great
potential
rational
design
next‐generation
materials.
integration
not
promises
unlock
novel
compounds
but
also
provides
new
opportunities
deepen
our
phenomena.
