-
Gemini: A Family of Highly Capable Multimodal Models
Authors:
Gemini Team,
Rohan Anil,
Sebastian Borgeaud,
Jean-Baptiste Alayrac,
Jiahui Yu,
Radu Soricut,
Johan Schalkwyk,
Andrew M. Dai,
Anja Hauth,
Katie Millican,
David Silver,
Melvin Johnson,
Ioannis Antonoglou,
Julian Schrittwieser,
Amelia Glaese,
Jilin Chen,
Emily Pitler,
Timothy Lillicrap,
Angeliki Lazaridou,
Orhan Firat,
James Molloy,
Michael Isard,
Paul R. Barham,
Tom Hennigan,
Benjamin Lee
, et al. (1325 additional authors not shown)
Abstract:
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultr…
▽ More
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.
△ Less
Submitted 17 June, 2024; v1 submitted 18 December, 2023;
originally announced December 2023.
-
Superconducting and Topological Properties in Centrosymmetric PbTaS2 Single Crystals
Authors:
J. J. Gao,
J. G. Si,
X. Luo,
J. Yan,
Z. Z. Jiang,
W. Wang,
C. Q. Xu,
X. F. Xu,
P. Tong,
W. H. Song,
X. B. Zhu,
W. J. Lu,
Y. P. Sun
Abstract:
We report the superconductivity of PbTaS2 single crystals with the centrosymmetric structure. The systematic measurements of magnetization, electric transport and specific heat indicate that PbTaS2 is a weakly coupled type-II superconductor with transition temperature Tc = 2.6 K. Furthermore, the band structure calculations predicted four nodal lines near the Fermi energy with drumhead-like surfac…
▽ More
We report the superconductivity of PbTaS2 single crystals with the centrosymmetric structure. The systematic measurements of magnetization, electric transport and specific heat indicate that PbTaS2 is a weakly coupled type-II superconductor with transition temperature Tc = 2.6 K. Furthermore, the band structure calculations predicted four nodal lines near the Fermi energy with drumhead-like surface states, suggesting centrosymmetric PbTaS2 is a candidate of topological nodal line semimetals. These results demonstrate that PbTaS2 may open up another avenue for further exploring the properties of superconductivity and topological nodal-line states.
△ Less
Submitted 21 May, 2020;
originally announced May 2020.
-
ParasNet: Fast Parasites Detection with Neural Networks
Authors:
X. F. Xu,
S. Talbot,
T. Selvaraja
Abstract:
Deep learning has dramatically improved the performance in many application areas such as image classification, object detection, speech recognition, drug discovery and etc since 2012. Where deep learning algorithms promise to discover the intricate hidden information inside the data by leveraging the large dataset, advanced model and computing power. Although deep learning techniques show medical…
▽ More
Deep learning has dramatically improved the performance in many application areas such as image classification, object detection, speech recognition, drug discovery and etc since 2012. Where deep learning algorithms promise to discover the intricate hidden information inside the data by leveraging the large dataset, advanced model and computing power. Although deep learning techniques show medical expert level performance in a lot of medical applications, but some of the applications are still not explored or under explored due to the variation of the species. In this work, we studied the bright field based cell level Cryptosporidium and Giardia detection in the drink water with deep learning. Our experimental demonstrates that the new developed deep learning-based algorithm surpassed the handcrafted SVM based algorithm with above 97 percentage in accuracy and 700+fps in speed on embedded Jetson TX2 platform. Our research will lead to real-time and high accuracy label-free cell level Cryptosporidium and Giardia detection system in the future.
△ Less
Submitted 24 March, 2020; v1 submitted 26 February, 2020;
originally announced February 2020.
-
Direct Evidence of Spontaneous Abrikosov Vortex State in Ferromagnetic Superconductor EuFe$_2$(As$_{1-x}$P$_x$)$_2$ with $x=0.21$
Authors:
L. Ya. Vinnikov,
I. S. Veshchunov,
S. Yu. Grebenchuk,
D. S. Baranov,
V. S. Stolyarov,
V. V. Dremov,
N. Zhou,
Z. X. Shi,
X. F. Xu,
S. Pyon,
Yue Sun,
Wenhe Jiao,
Guanghan Cao,
A. A. Golubov,
D. Roditchev,
A. I. Buzdin,
T. Tamegai
Abstract:
Using low-temperature Magnetic Force Microscopy (MFM) we provide direct experimental evidence for spontaneous vortex phase (SVP) formation in EuFe$_2$(As$_{0.79}$P$_{0.21}$)$_2$ single crystal with the superconducting $T^{\rm 0}_{\rm SC}=23.6$~K and ferromagnetic $T_{\rm FM}\sim17.7$~K transition temperatures. Spontaneous vortex-antivortex (V-AV) pairs are imaged in the vicinity of $T_{\rm FM}$. A…
▽ More
Using low-temperature Magnetic Force Microscopy (MFM) we provide direct experimental evidence for spontaneous vortex phase (SVP) formation in EuFe$_2$(As$_{0.79}$P$_{0.21}$)$_2$ single crystal with the superconducting $T^{\rm 0}_{\rm SC}=23.6$~K and ferromagnetic $T_{\rm FM}\sim17.7$~K transition temperatures. Spontaneous vortex-antivortex (V-AV) pairs are imaged in the vicinity of $T_{\rm FM}$. Also, upon cooling cycle near $T_{\rm FM}$ we observe the first-order transition from the short period domain structure, which appears in the Meissner state, into the long period domain structure with spontaneous vortices. It is the first experimental observation of this scenario in the ferromagnetic superconductors. Low-temperature phase is characterized by much larger domains in V-AV state and peculiar branched striped structures at the surface, which are typical for uniaxial ferromagnets with perpendicular magnetic anisotropy (PMA). The domain wall parameters at various temperatures are estimated.
△ Less
Submitted 29 September, 2017; v1 submitted 28 September, 2017;
originally announced September 2017.
-
Visualization of Magnetic Flux Structure in Phosphorus-Doped EuFe$_2$As$_2$ Single Crystals
Authors:
I. S. Veshchunov,
L. Ya. Vinnikov,
V. S. Stolyarov,
N. Zhou,
Z. X. Shi,
X. F. Xu,
S. Yu. Grebenchuk,
D. S. Baranov,
I. A. Golovchanskiy,
S. Pyon,
Yue Sun,
Wenhe Jiao,
Guanghan Cao,
T. Tamegai,
A. A. Golubov
Abstract:
Magnetic flux structure on the surface of EuFe$_2$(As$\rm_{1-x}$P$\rm_x$)$_2$ single crystals with nearly optimal phosphorus do** levels $x=0.20$, and $x=0.21$ is studied by low-temperature magnetic force microscopy and decoration with ferromagnetic nanoparticles. The studies are performed in a broad temperature range. It is shown that the single crystal with $x=0.21$ in the temperature range be…
▽ More
Magnetic flux structure on the surface of EuFe$_2$(As$\rm_{1-x}$P$\rm_x$)$_2$ single crystals with nearly optimal phosphorus do** levels $x=0.20$, and $x=0.21$ is studied by low-temperature magnetic force microscopy and decoration with ferromagnetic nanoparticles. The studies are performed in a broad temperature range. It is shown that the single crystal with $x=0.21$ in the temperature range between the critical temperatures $T_{\rm SC}=22$ K and $T_{\rm C}=17.7$ K of the superconducting and ferromagnetic phase transitions, respectively, has the vortex structure of a frozen magnetic flux, typical for type-II superconductors. The magnetic domain structure is observed in the superconducting state below $T_{\rm C}$. The nature of this structure is discussed.
△ Less
Submitted 7 March, 2017;
originally announced March 2017.
-
Magnetic phase diagram in the Co-rich side of LnCo1-xFexAsO (Ln=La, Sm) system
Authors:
Y. K. Li,
X. F. Xu,
C. Cao,
C. Y. Shen,
Y. K. Luo,
Q. Tao,
X. Lin,
L. Zhang,
G. H. Cao,
Z. A. Xu
Abstract:
The magnetic phase diagram has been mapped out via the measurements of electronic resistivity, magnetization and specific heat in the cobalt-based layered LnCo1-xFexAsO (Ln=La, Sm) compounds. The ferromagnetic (FM) transition at 63 K for LaCoAsO is rapidly suppressed upon Fe do**, and ultimately disappears around x=0.3 in the LaCo1-xFexAsO system. When La is replaced by magnetic rare earth eleme…
▽ More
The magnetic phase diagram has been mapped out via the measurements of electronic resistivity, magnetization and specific heat in the cobalt-based layered LnCo1-xFexAsO (Ln=La, Sm) compounds. The ferromagnetic (FM) transition at 63 K for LaCoAsO is rapidly suppressed upon Fe do**, and ultimately disappears around x=0.3 in the LaCo1-xFexAsO system. When La is replaced by magnetic rare earth element Sm, the 3d electrons first undergo a FM transition at Tc = 75 K, followed by an antiferromagnetic (AFM) transition at a lower temperature TN1 = 45 K. With partial Fe do** on the Co site, both FM (Tc) and AFM (TN1) transition temperatures are significantly suppressed, and finally approach zero kelvin at x = 0.3 and 0.2, respectively. Meanwhile, a third magnetic transition at TN2 = 5.6 K for SmCoAsO, associated with the AFM order of the Sm3+ 4f-oments, is uncovered and TN2 is found to be almost robust against the small Fe-do**. These results suggest that the 4f electrons of Sm3+ have an important effect on the magnetic behavior of 3d electrons in the 1111 type Co-based LnCo1-xFexAsO systems. In contrast, the magnetism of the f-electrons is relatively unaffected by the variation of the 3d electrons. The rich magnetic phase diagram in the Co-rich side of the LnCo1-xFexAsO system, therefore, is established.
△ Less
Submitted 10 January, 2013; v1 submitted 23 August, 2012;
originally announced August 2012.
-
Band-dependent normal-state coherence in Sr$_{2}$RuO$_{4}$: Evidence from Nernst effect and thermopower measurements
Authors:
X. F. Xu,
Z. A. Xu,
T. J. Liu,
D. Fobes,
Z. Q. Mao,
J. L. Luo,
Y. Liu
Abstract:
We present the first measurement on Nernst effect in the normal state of odd-parity, spin-triplet superconductor Sr$_{2}$RuO$_{4}$. Below 100 K, the Nernst signal was found to be negative, large, and, as a function of magnetic field, nonlinear. Its magnitude increases with the decreasing temperature until reaching a maximum around $T^*$ $\approx$ 20 - 25 K, below which it starts to decrease line…
▽ More
We present the first measurement on Nernst effect in the normal state of odd-parity, spin-triplet superconductor Sr$_{2}$RuO$_{4}$. Below 100 K, the Nernst signal was found to be negative, large, and, as a function of magnetic field, nonlinear. Its magnitude increases with the decreasing temperature until reaching a maximum around $T^*$ $\approx$ 20 - 25 K, below which it starts to decrease linearly as a function of temperature. The large value of the Nernst signal appears to be related to the multiband nature of the normal state and the nonlinearity to band-dependent magnetic fluctuation in Sr$_{2}$RuO$_{4}$. We argue that the sharp decrease in Nernst signal below $T^*$ is due to the suppression of quasiparticle scattering and the emergence of band-dependent coherence in the normal state. The observation of a sharp kink in the temperature dependent thermopower around $T^*$ and a sharp drop of Hall angle at low temperatures provide additional support to this picture.
△ Less
Submitted 5 July, 2008; v1 submitted 22 January, 2008;
originally announced January 2008.
