arXiv subjects

Cong Yu

Publications and source records attributed to Cong Yu.

At least 55 records · Page 3Linked to original sources

arXiv2023

Long-term Orbital Period Variation of Hot Jupiters from Transiting Time Analysis using TESS Survey Data

Many hot Jupiters may experience orbital decays, which are manifested as long-term transit timing variations. We have analyzed 7068 transits from the Transiting Exoplanet Survey Satellite (TESS) for a sample of 326 hot Jupiters. These new mid-transit time data allow us to update ephemerides for these systems. By combining the new TESS transit timing data with archival data, we search for possible long-term orbital period variations in these hot Jupiters using a linear and a quadratic ephemeris model. We identified 26 candidates that exhibit possible long-term orbital period variations, including 18 candidates with decreasing orbital periods and 8 candidates with increasing orbital periods. Among them, 12 candidates have failed in our leave-one-out cross-validation (LOOCV) test and thus should be considered as marginal candidates. In addition to tidal interaction, alternative mechanisms such as apsidal precession, Rømer effect, and Applegate effect could also contribute to the observed period variations. The ephemerides derived in this work are useful for scheduling follow-up observations for these hot Jupiters in the future. The Python code used to generate the ephemerides is made available online.

astro-ph.EP↗

arXiv2023

Are Large Language Models Really Robust to Word-Level Perturbations?

The swift advancement in the scales and capabilities of Large Language Models (LLMs) positions them as promising tools for a variety of downstream tasks. In addition to the pursuit of better performance and the avoidance of violent feedback on a certain prompt, to ensure the responsibility of the LLM, much attention is drawn to the robustness of LLMs. However, existing evaluation methods mostly rely on traditional question answering datasets with predefined supervised labels, which do not align with the superior generation capabilities of contemporary LLMs. To address this issue, we propose a novel rational evaluation approach that leverages pre-trained reward models as diagnostic tools to evaluate the longer conversation generated from more challenging open questions by LLMs, which we refer to as the Reward Model for Reasonable Robustness Evaluation (TREvaL). Longer conversations manifest the comprehensive grasp of language models in terms of their proficiency in understanding questions, a capability not entirely encompassed by individual words or letters, which may exhibit oversimplification and inherent biases. Our extensive empirical experiments demonstrate that TREvaL provides an innovative method for evaluating the robustness of an LLM. Furthermore, our results demonstrate that LLMs frequently exhibit vulnerability to word-level perturbations that are commonplace in daily language usage. Notably, we are surprised to discover that robustness tends to decrease as fine-tuning (SFT and RLHF) is conducted. The code of TREval is available in https://github.com/Harry-mic/TREvaL.

cs.CL↗

arXiv2023

Ohmic dissipation during the formation of super-Earth

Super-Earth population, as one of the representatives of exoplanets, plays an important role in constraining the planet formation theories. According to the prediction from core-accretion models, super-Earths should be rare because their masses are in the range of the critical mass above which they would grow to be gas giants by runaway gas accretion. In this work, we investigate the effect of ohmic dissipation on the planetary thermal structure and cooling contraction as planets accrete gas from their surrounding disks. We find that the extra heating energy from Ohmic heating deposited into planetary envelopes can push the planetary radiative-convective boundaries inward and prevent the planets from cooling, and can even halt accretion. We explore parameter space to study the dependence of cooling timescale on the input parameters of the ohmic-dissipation model. Numerical results show that gas accretion can be halted before runaway gas accretion and the envelope mass is only several percent of planetary core mass for some parameter sets. Our results suggest that ohmic dissipation is a potential mechanism to delay the gas accretion and promote the formation of super-Earths. Future observations may help to constrain the importance of ohmic dissipation on the super-Earth formation.

astro-ph.EP↗

arXiv2022

RFPose-OT: RF-Based 3D Human Pose Estimation via Optimal Transport Theory

This paper introduces a novel framework, i.e., RFPose-OT, to enable the 3D human pose estimation from Radio Frequency (RF) signals. Different from existing methods that predict human poses from RF signals on the signal level directly, we consider the structure difference between the RF signals and the human poses, propose to transform the RF signals to the pose domain on the feature level based on Optimal Transport (OT) theory, and generate human poses from the transformed features. To evaluate RFPose-OT, we build a radio system and a multi-view camera system to acquire the RF signal data and the ground-truth human poses. The experimental results in basic indoor environment, occlusion indoor environment, and outdoor environment, all demonstrate that RFPose-OT can predict 3D human poses with higher precision than the state-of-the-art methods.

cs.CV↗

arXiv2022

Boundary Layers of Circumplanetary Disks around Spinning Planets I. Effects of Rossby Waves

Gas giant planets are believed to accrete from their circumplanetary disks (CPDs). The CPDs usually involve accretion through the boundary layer (BL) in the vicinity of planets. Prior studies have concentrated on the BL of non-spinning planets. We investigate the influence of planetary spin on the wave behaviors within the BL. The rotation profile in such BLs would show sharp transition from the rigid rotation to the Keplerian rotation. We examine the angular momentum transport in these BL in terms of linear perturbation analysis. We find that the global inertia-acoustic mode associated with spinning planets would give rise to the inflow of angular momentum and the accretion of gas. In this work, we identify a new kind of global mode, namely the Rossby mode. The Rossby mode can lead to the outflow of angular momentum and the decretion of gas from a spinning planet. The Rossby mode provide a negative feedback that regulates the planetary spin and mass. We compare the growth rate of the two modes as a function of the width of BL, the Mach number and the spin rate of planet. Our results reveal the underlying hydrodynamic mechanism of terminal spins and asymptotic mass of the giant planets.

astro-ph.EP↗

arXiv2022

Pulsar Glitches: A Review

$\sim 6\%$ of all known pulsars have been observed to exhibit sudden spin-up events, known as glitches. For more than fifty years, these phenomena have played an important role in helping to understand pulsar (astro)physics. Based on the review of pulsar glitches search method, the progress made in observations in recent years is summarized, including the achievements obtained by Chinese telescopes. Glitching pulsars demonstrate great diversity of behaviours, which can be broadly classified into four categories: normal glitches, slow glitches, glitches with delayed spin-ups, and anti-glitches. The main models of glitches that have been proposed are reviewed and their implications for neutron star structure are critically examined regarding our current understanding. Furthermore, the correlations between glitches and emission changes, which suggest that magnetospheric state-change is linked to the pulsar-intrinsic processes, are also described and discussed in some detail.

astro-ph.HE↗

arXiv2022

Scaling Multimodal Pre-Training via Cross-Modality Gradient Harmonization

Self-supervised pre-training recently demonstrates success on large-scale multimodal data, and state-of-the-art contrastive learning methods often enforce the feature consistency from cross-modality inputs, such as video/audio or video/text pairs. Despite its convenience to formulate and leverage in practice, such cross-modality alignment (CMA) is only a weak and noisy supervision, since two modalities can be semantically misaligned even they are temporally aligned. For example, even in the commonly adopted instructional videos, a speaker can sometimes refer to something that is not visually present in the current frame; and the semantic misalignment would only be more unpredictable for the raw videos from the internet. We conjecture that might cause conflicts and biases among modalities, and may hence prohibit CMA from scaling up to training with larger and more heterogeneous data. This paper first verifies our conjecture by observing that, even in the latest VATT pre-training using only instructional videos, there exist strong gradient conflicts between different CMA losses within the same video, audio, text triplet, indicating them as the noisy source of supervision. We then propose to harmonize such gradients, via two techniques: (i) cross-modality gradient realignment: modifying different CMA loss gradients for each sample triplet, so that their gradient directions are more aligned; and (ii) gradient-based curriculum learning: leveraging the gradient conflict information on an indicator of sample noisiness, to develop a curriculum learning strategy to prioritize training on less noisy sample triplets. Applying those techniques to pre-training VATT on the HowTo100M dataset, we consistently improve its performance on different downstream tasks. Moreover, we are able to scale VATT pre-training to more complicated non-narrative Youtube8M dataset to further improve the state-of-the-arts.

cs.CV↗

arXiv2022

ET White Paper: To Find the First Earth 2.0

We propose to develop a wide-field and ultra-high-precision photometric survey mission, temporarily named "Earth 2.0 (ET)". This mission is designed to measure, for the first time, the occurrence rate and the orbital distributions of Earth-sized planets. ET consists of seven 30cm telescopes, to be launched to the Earth-Sun's L2 point. Six of these are transit telescopes with a field of view of 500 square degrees. Staring in the direction that encompasses the original Kepler field for four continuous years, this monitoring will return tens of thousands of transiting planets, including the elusive Earth twins orbiting solar-type stars. The seventh telescope is a 30cm microlensing telescope that will monitor an area of 4 square degrees toward the galactic bulge. This, combined with simultaneous ground-based KMTNet observations, will measure masses for hundreds of long-period and free-floating planets. Together, the transit and the microlensing telescopes will revolutionize our understandings of terrestrial planets across a large swath of orbital distances and free space. In addition, the survey data will also facilitate studies in the fields of asteroseismology, Galactic archeology, time-domain sciences, and black holes in binaries.

astro-ph.IM↗

arXiv2022

Rossby Wave Instabilities of Protoplanetary Discs with Cooling

Rossby wave instabilities (RWIs) usually lead to nonaxisymmetric vortices in protoplanetary discs and some observed sub-structures of these discs can be well explained by RWIs. We explore how the cooling influences the growth rate of unstable RWI modes in terms of the linear perturbation analysis. The cooling associated with the energy equation is treated in two different ways. The first one we adopt is a simple cooling law. The perturbed thermal state relaxes to the initial thermal state on a prescribed cooling timescale. In the second, we treat the cooling as a thermal diffusion process. The difference in the growth rate between the adiabatic and isothermal modes becomes more pronounced for discs with smaller sound speed. For the simple cooling law, the growth rates of unstable modes monotonically decrease with the shorter cooling timescale in barotropic discs. But the dependence of growth rate with the cooling timescale becomes non-monotonic in non-baratopic discs. The RWI might even be enhanced in non-barotropic discs during the transition from the adiabatic state to the isothermal state. When the cooling is treated as the thermal diffusion, even in barotropic disc, the variation of growth rate with thermal diffusivity becomes non-monotonic. Further more, a maximum growth rate may appear with an appropriate value of thermal diffusivity. The angular momentum flux (AMF) is investigated to understand the angular momentum transport by RWI with cooling.

astro-ph.EP↗

arXiv2022

All Birds with One Stone: Multi-task Text Classification for Efficient Inference with One Forward Pass

Multi-Task Learning (MTL) models have shown their robustness, effectiveness, and efficiency for transferring learned knowledge across tasks. In real industrial applications such as web content classification, multiple classification tasks are predicted from the same input text such as a web article. However, at the serving time, the existing multitask transformer models such as prompt or adaptor based approaches need to conduct N forward passes for N tasks with O(N) computation cost. To tackle this problem, we propose a scalable method that can achieve stronger performance with close to O(1) computation cost via only one forward pass. To illustrate real application usage, we release a multitask dataset on news topic and style classification. Our experiments show that our proposed method outperforms strong baselines on both the GLUE benchmark and our news dataset. Our code and dataset are publicly available at https://bit.ly/mtop-code.

cs.CL↗

arXiv2022

Effects of Self-gravity on Mass-loss of the Post-impact Super-Earths

Kepler's observations show most of the exoplanets are super-Earths. The formation of super-Earth is generally related to the atmospheric mass loss that is crucial in the planetary structure and evolution. The shock driven by the giant impact will heat the planet, resulting in the atmosphere escape. We focus on whether self-gravity changes the efficiency of mass loss. Without self-gravity, if the impactor mass is comparable to the envelope mass, there is a significant mass-loss. The radiative-convective boundary will shift inward by self-gravity. As the temperature and envelope mass increase, the situation becomes more prominent, resulting in a heavier envelope. Therefore, the impactor mass will increase to motivate the significant mass loss, as the self-gravity is included. With the increase of envelope mass, the self-gravity is particularly important.

astro-ph.EP↗

arXiv2022

Image Steganography based on Style Transfer

Image steganography is the art and science of using images as cover for covert communications. With the development of neural networks, traditional image steganography is more likely to be detected by deep learning-based steganalysis. To improve upon this, we propose image steganography network based on style transfer, and the embedding of secret messages can be disguised as image stylization. We embed secret information while transforming the content image style. In latent space, the secret information is integrated into the latent representation of the cover image to generate the stego images, which are indistinguishable from normal stylized images. It is an end-to-end unsupervised model without pre-training. Extensive experiments on the benchmark dataset demonstrate the reliability, quality and security of stego images generated by our steganographic network.

cs.CV↗

arXiv2022

Training ELECTRA Augmented with Multi-word Selection

Pre-trained text encoders such as BERT and its variants have recently achieved state-of-the-art performances on many NLP tasks. While being effective, these pre-training methods typically demand massive computation resources. To accelerate pre-training, ELECTRA trains a discriminator that predicts whether each input token is replaced by a generator. However, this new task, as a binary classification, is less semantically informative. In this study, we present a new text encoder pre-training method that improves ELECTRA based on multi-task learning. Specifically, we train the discriminator to simultaneously detect replaced tokens and select original tokens from candidate sets. We further develop two techniques to effectively combine all pre-training tasks: (1) using attention-based networks for task-specific heads, and (2) sharing bottom layers of the generator and the discriminator. Extensive experiments on GLUE and SQuAD datasets demonstrate both the effectiveness and the efficiency of our proposed method.

cs.CL↗

arXiv2022

Charformer: Fast Character Transformers via Gradient-based Subword Tokenization

State-of-the-art models in natural language processing rely on separate rigid subword tokenization algorithms, which limit their generalization ability and adaptation to new settings. In this paper, we propose a new model inductive bias that learns a subword tokenization end-to-end as part of the model. To this end, we introduce a soft gradient-based subword tokenization module (GBST) that automatically learns latent subword representations from characters in a data-driven fashion. Concretely, GBST enumerates candidate subword blocks and learns to score them in a position-wise fashion using a block scoring network. We additionally introduce Charformer, a deep Transformer model that integrates GBST and operates on the byte level. Via extensive experiments on English GLUE, multilingual, and noisy text datasets, we show that Charformer outperforms a series of competitive byte-level baselines while generally performing on par and sometimes outperforming subword-based models. Additionally, Charformer is fast, improving the speed of both vanilla byte-level and subword-level Transformers by 28%-100% while maintaining competitive quality. We believe this work paves the way for highly performant token-free models that are trained completely end-to-end.

cs.CL↗

arXiv2022

RFMask: A Simple Baseline for Human Silhouette Segmentation with Radio Signals

Human silhouette segmentation, which is originally defined in computer vision, has achieved promising results for understanding human activities. However, the physical limitation makes existing systems based on optical cameras suffer from severe performance degradation under low illumination, smoke, and/or opaque obstruction conditions. To overcome such limitations, in this paper, we propose to utilize the radio signals, which can traverse obstacles and are unaffected by the lighting conditions to achieve silhouette segmentation. The proposed RFMask framework is composed of three modules. It first transforms RF signals captured by millimeter wave radar on two planes into spatial domain and suppress interference with the signal processing module. Then, it locates human reflections on RF frames and extract features from surrounding signals with human detection module. Finally, the extracted features from RF frames are aggregated with an attention based mask generation module. To verify our proposed framework, we collect a dataset containing 804,760 radio frames and 402,380 camera frames with human activities under various scenes. Experimental results show that the proposed framework can achieve impressive human silhouette segmentation even under the challenging scenarios(such as low light and occlusion scenarios) where traditional optical-camera-based methods fail. To the best of our knowledge, this is the first investigation towards segmenting human silhouette based on millimeter wave signals. We hope that our work can serve as a baseline and inspire further research that perform vision tasks with radio signals. The dataset and codes will be made in public.

cs.CV↗

arXiv2021

The Critical Core Mass of Rotating Planets

The gravitational harmonics measured from Juno and Cassini spacecrafts help us to specify the internal structure and chemical elements of Jupiter and Saturn, respectively. However, we still do not know much about the impact of rotation on the planetary internal structure as well as their formation. The centrifugal force induced by rotation deforms the planetary shape and partially counteracts the gravitational force. Thus, rotation will affect the critical core mass of the exoplanet. Once the atmospheric mass becomes comparable to the critical core mass, the planet will enter the runaway accretion phase and becomes a gas giant. We have confirmed that the critical core masses of rotating planets depend on the stiffness of the polytrope, the outer boundary conditions, and the thickness of the isothermal layer. The critical core mass with Bondi boundary condition is determined by the surface properties. The critical core mass of a rotating planet will increase with the core gravity (i.e., the innermost density). For the Hill boundary condition, the soft polytrope shares the same properties as planets with Bondi boundary condition. Since the total mass for planets with Hill boundary condition increases with the decrease of the polytropic index, higher core gravity is required for rotating planets. As a result, the critical core mass in the stiff Hill model sharply increases. The rotation effects become more important when the radiative and convective regions coexist. Besides, the critical core mass of planets with Hill (Bondi) boundary increases noticeably as the radiative layer becomes thinner (thicker).

astro-ph.EP↗

arXiv2021

RFGAN: RF-Based Human Synthesis

This paper demonstrates human synthesis based on the Radio Frequency (RF) signals, which leverages the fact that RF signals can record human movements with the signal reflections off the human body. Different from existing RF sensing works that can only perceive humans roughly, this paper aims to generate fine-grained optical human images by introducing a novel cross-modal RFGAN model. Specifically, we first build a radio system equipped with horizontal and vertical antenna arrays to transceive RF signals. Since the reflected RF signals are processed as obscure signal projection heatmaps on the horizontal and vertical planes, we design a RF-Extractor with RNN in RFGAN for RF heatmap encoding and combining to obtain the human activity information. Then we inject the information extracted by the RF-Extractor and RNN as the condition into GAN using the proposed RF-based adaptive normalizations. Finally, we train the whole model in an end-to-end manner. To evaluate our proposed model, we create two cross-modal datasets (RF-Walk & RF-Activity) that contain thousands of optical human activity frames and corresponding RF signals. Experimental results show that the RFGAN can generate target human activity frames using RF signals. To the best of our knowledge, this is the first work to generate optical images based on RF signals.

cs.MM↗

arXiv2021

In situ formation of super-Earth/sub-Neptune driven by the planetary rotation

Kepler's observation shows that many of the detected planets are super-Earths. They are inside a range of critical masses overlapping the core masses (2-20 $M_{\bigoplus}$), which would trigger the runaway accretion and develop the gas giants. Thus, super-Earths/sub-Neptunes can be formed by restraining runaway growth of gaseous envelopes. We assess the effect of planetary rotation in delaying the mass growth. The centrifugal force, induced by spin, will offset a part of the gravitational force and deform the planet. Tracking the change in structure, we find that the temperature at the radiative-convective boundary (RCB) is approximate to the boundary temperature. Since rotation reduces the radiation energy densities in the convective and radiative layers, RCB will penetrate deeper. The cooling luminosity would decrease. Under this condition, the evolutionary timescale can exceed the disk lifetime (10 Myr), and a super-Earth/sub-Neptune could be formed after undergoing additional mass loss processes. In the dusty atmosphere, even a lower angular velocity can also promote a super-Earth/sub-Neptune forming. Therefore, we conclude that rotation can slow down the planet's cooling and then promote a super-Earth/sub-Neptune forming.

astro-ph.EP↗