Search

Kinship face synthesis is a challenging problem due to the scarcity and low quality of the available kinship data. Existing methods often struggle to generate descendants with both high diversity and fidelity while precisely controlling facial attributes such as age and gender. To address these issues, we propose the Style Latent Diffusion Transformer (StyleDiT), a novel framework that integrates the strengths of StyleGAN with the diffusion model to generate high-quality and diverse kinship faces. In this framework, the rich facial priors of StyleGAN enable fine-grained attribute control, while our conditional diffusion model is used to sample a StyleGAN latent aligned with the kinship relationship of conditioning images by utilizing the advantage of modeling complex kinship relationship distribution. StyleGAN then handles latent decoding for final face generation. Additionally, we introduce the Relational Trait Guidance (RTG) mechanism, enabling independent control of influencing conditions, such as each parent's facial image. RTG also enables a fine-grained adjustment between the diversity and fidelity in synthesized faces. Furthermore, we extend the application to an unexplored domain: predicting a partner's facial images using a child's image and one parent's image within the same framework. Extensive experiments demonstrate that our StyleDiT outperforms existing methods by striking an excellent balance between generating diverse and high-fidelity kinship faces.

In competitive games, strength ratings like Elo are widely used to quantify player skill and support matchmaking by accounting for skill disparities better than simple win rate statistics. However, scalar ratings cannot handle complex intransitive relationships, such as counter strategies seen in Rock-Paper-Scissors. To address this, recent work introduced Neural Rating Table and Neural Counter Table, which combine scalar ratings with discrete counter categories to model intransitivity. While effective, these methods rely on neural network training and cannot perform real-time updates. In this paper, we propose an online update algorithm that extends Elo principles to incorporate real-time learning of counter categories. Our method dynamically adjusts both ratings and counter relationships after each match, preserving the explainability of scalar ratings while addressing intransitivity. Experiments on zero-sum competitive games demonstrate its practicality, particularly in scenarios without complex team compositions.

Ultra-light axions are hypothetical scalar particles that influence the evolution of large-scale structures of the Universe. Depending on their mass, they can potentially be part of the dark matter component of the Universe, as candidates commonly referred to as fuzzy dark matter. While strong constraints have been established for pure fuzzy dark matter models, the more general scenario where ultra-light axions constitute only a fraction of the dark matter has been limited to a few observational probes. In this work, we use the galaxy cluster number counts obtained from the first All-Sky Survey (eRASS1) of the SRG/eROSITA mission together with gravitational weak lensing data from the Dark Energy Survey, the Kilo-Degree Survey, and the Hyper Suprime-Cam, to constrain the fraction of ultra-light axions in the mass range $10^{-32}$ eV to $10^{-24}$ eV. We put upper bounds on the ultra-light axion relic density in independent logarithmic axion mass bins by performing a full cosmological parameter inference. We find an exclusion region in the intermediate ultra-light axion mass regime with the tightest bounds reported so far in the mass bins around $m_\mathrm{a}=10^{-27}$ eV with $\Omega_\mathrm{a} < 0.0036$ and $m_\mathrm{a}=10^{-26}$ eV with $\Omega_\mathrm{a} < 0.0084$, both at 95% confidence level. When combining with CMB probes, these bounds are tightened to $\Omega_\mathrm{a} < 0.0030$ in the $m_\mathrm{a}=10^{27}$ eV mass bin and $\Omega_\mathrm{a} < 0.0058$ in the $m_\mathrm{a}=10^{-26}$ eV mass bin, both at 95% confidence level. This is the first time that constraints on ultra-light axions have been obtained using the growth of structure measured by galaxy cluster number counts. These results pave the way for large surveys, which can be utilized to obtain tight constraints on the mass and relic density of ultra-light axions with better theoretical modeling of the abundance of halos., Comment: 16 pages, 11 figures, submitted to A&A

Large Language Models (LLMs) have seen widespread applications across various domains due to their growing ability to process diverse types of input data, including text, audio, image and video. While LLMs have demonstrated outstanding performance in understanding and generating contexts for different scenarios, they are vulnerable to prompt-based attacks, which are mostly via text input. In this paper, we introduce the first voice-based jailbreak attack against multimodal LLMs, termed as Flanking Attack, which can process different types of input simultaneously towards the multimodal LLMs. Our work is motivated by recent advancements in monolingual voice-driven large language models, which have introduced new attack surfaces beyond traditional text-based vulnerabilities for LLMs. To investigate these risks, we examine the frontier multimodal LLMs, which can be accessed via different types of inputs such as audio input, focusing on how adversarial prompts can bypass its defense mechanisms. We propose a novel strategy, in which the disallowed prompt is flanked by benign, narrative-driven prompts. It is integrated in the Flanking Attack which attempts to humanizes the interaction context and execute the attack through a fictional setting. To better evaluate the attack performance, we present a semi-automated self-assessment framework for policy violation detection. We demonstrate that Flank Attack is capable of manipulating state-of-the-art LLMs into generating misaligned and forbidden outputs, which achieves an average attack success rate ranging from 0.67 to 0.93 across seven forbidden scenarios. These findings highlight both the potency of prompt-based obfuscation in voice-enabled contexts and the limitations of current LLMs' moderation safeguards and the urgent need for advanced defense strategies to address the challenges posed by evolving, context-rich attacks.

Recent advances in deep learning have established Transformer architectures as the predominant modeling paradigm. Central to the success of Transformers is the self-attention mechanism, which scores the similarity between query and key matrices to modulate a value matrix. This operation bears striking similarities to digraph convolution, prompting an investigation into whether digraph convolution could serve as an alternative to self-attention. In this study, we formalize this concept by introducing a synthetic unitary digraph convolution based on the digraph Fourier transform. The resulting model, which we term Converter, effectively converts a Transformer into a Directed Graph Neural Network (DGNN) form. We have tested Converter on Long-Range Arena benchmark, long document classification, and DNA sequence-based taxonomy classification. Our experimental results demonstrate that Converter achieves superior performance while maintaining computational efficiency and architectural simplicity, which establishes it as a lightweight yet powerful Transformer variant., Comment: 21 pages, 3 figures, and 8 tables

Functionalities in crystalline materials are determined by 3-dimensional collective interactions of atoms. The confinement of dimensionality in condensed matter provides an exotic research direction to understand the interaction of atoms, thus can be used to tailor or create new functionalities in material systems. In this study, a 2-dimensional transition metal oxide monolayer is constructed inside complex oxide heterostructures based on the theoretical predictions. The electrostatic boundary conditions of oxide monolayer in the heterostructure is carefully designed to tune the chemical, electronic, and magnetic states of oxide monolayer. The challenge of characterizing such an oxide monolayer is overcome by a combination of transmission electron microscopy, x-ray absorption spectroscopy, cross-sectional scanning tunneling microscopy, and electrical transport measurements. An intriguing metal-insulator transition associated with a magnetic transition is discovered in the MnO2 monolayer. This study paves a new route to understand the confinement of dimensionality and explore new intriguing phenomena in condensed matters., Comment: 23 pages, 6 figures

Unlike broken time-reversal symmetric (TRS) systems with a defined Chern number, directly measuring the bulk $Z_{2}$ invariant and Berry curvature (if nonzero) in topological insulators and their higher-order topological families remains an unsolved problem. Here, based on the refined trace-determinant inequality (TDI) involving the trace and determinant of the quantum metric and maximal Berry curvature (MBC), we establish an optical bound on the MBC for two-dimensional TRS insulators. By utilizing (certain energy range of) experimental data on optical conductivity, one can identify topological signatures using the integrated optical bound. We illustrate our approach using three representative topological models: the Kane-Mele model, mirror-protected insulator, and quadrupole insulator. We find that the resulting momentum integration of the refined TDI provides an intrinsic relationship between quantum volume (QV) variations and topological phase transitions. We argue that double QV can be treated as an upper bound on the number of boundary states. Our findings offer a method for extracting the topological signatures of TRS insulators using optical conductivity data.

In endovascular surgery, the precise identification of catheters and guidewires in X-ray images is essential for reducing intervention risks. However, accurately segmenting catheter and guidewire structures is challenging due to the limited availability of labeled data. Foundation models offer a promising solution by enabling the collection of similar domain data to train models whose weights can be fine-tuned for downstream tasks. Nonetheless, large-scale data collection for training is constrained by the necessity of maintaining patient privacy. This paper proposes a new method to train a foundation model in a decentralized federated learning setting for endovascular intervention. To ensure the feasibility of the training, we tackle the unseen data issue using differentiable Earth Mover's Distance within a knowledge distillation framework. Once trained, our foundation model's weights provide valuable initialization for downstream tasks, thereby enhancing task-specific performance. Intensive experiments show that our approach achieves new state-of-the-art results, contributing to advancements in endovascular intervention and robotic-assisted endovascular surgery, while addressing the critical issue of data sharing in the medical domain., Comment: 8 pages. Accepted to ICRA 2025

This paper investigates reconfigurable intelligent surface (RIS)-assisted unmanned aerial vehicle (UAV) downlink networks with fluid antennas (FA), where RIS enables non-line-of-sight (NLoS) transmissions. Moreover, the FA is equipped on the UAV offering dynamic antenna position adjustment, enhancing spatial diversity besides UAV deployment. We aim at total downlink rate maximization while ensuring minimum user rate requirement. We consider joint optimization of active UAV beamforming, passive RIS beamforming, UAV deployment and FA position adjustment. To address the complex problem, we propose beamfomring for RIS/UAV and FA-UAV deployment (BRAUD) scheme by employing alternative optimization, successive convex approximation (SCA) and sequential rank-one constraint relaxation (SROCR) method for the decomposed subproblems. Simulation results demonstrate the effectiveness of RIS-FA-UAV, achieving the highest rate among existing architectures without FA/UAV/RIS deployment and without proper beamforming. Moreover, BRAUD achieves the highest rate among benchmarks of drop-rank method, heuristic optimizations and conventional zero-forcing beamforming as well as random method., Comment: accepted by IEEE WCL

Observations of massive supermassive black holes (SMBHs) in the early universe challenge existing black hole formation models. We propose that soliton cores in fuzzy dark matter (FDM) offer a potential solution to this timing problem. Our FDM cosmological zoom-in simulations confirm that for a particle mass $m_{\rm FDM}\sim 10^{-22}~{\rm eV}$, solitons are well developed at redshift $z \sim 7$ with masses of $\sim10^9~M_\odot$, comparable to the observed SMBHs. We then demonstrate using hydrodynamic simulations that, compared to cold dark matter, these high-$z$ massive FDM solitons with mass $M_s$ can provide additional gravitational potential to accrete gas and boost the Bondi accretion rate of a growing black hole seed with mass $M_{\rm BH}$ by up to two to four orders of magnitude, in the regime of efficient cooling and negligible radiation pressure. This accretion boosting mechanism is effective for $10^{-22}~{\rm eV} \lesssim m_{\rm FDM} \lesssim 10^{-20}~{\rm eV}$ and potentially beyond as long as $M_s > M_{\rm BH}$. The simulation code GAMER is accessible at https://github.com/gamer-project/gamer., Comment: 9 pages, 4 figures, 1 table. Accepted for publication in Physical Review Letters

Recently, diffusion-based blind super-resolution (SR) methods have shown great ability to generate high-resolution images with abundant high-frequency detail, but the detail is often achieved at the expense of fidelity. Meanwhile, another line of research focusing on rectifying the reverse process of diffusion models (i.e., diffusion guidance), has demonstrated the power to generate high-fidelity results for non-blind SR. However, these methods rely on known degradation kernels, making them difficult to apply to blind SR. To address these issues, we present DADiff in this paper. DADiff incorporates degradation-aware models into the diffusion guidance framework, eliminating the need to know degradation kernels. Additionally, we propose two novel techniques: input perturbation and guidance scalar, to further improve our performance. Extensive experimental results show that our proposed method has superior performance over state-of-the-art methods on blind SR benchmarks., Comment: To appear in WACV 2025. Code is available at: https://github.com/ryanlu2240/DADiff

With the rapid advancement of codec-based speech generation (CoSG) systems, creating fake speech that mimics an individual's identity and spreads misinformation has become remarkably easy. Addressing the risks posed by such deepfake speech has attracted significant attention. However, most existing studies focus on detecting fake data generated by traditional speech generation models. Research on detecting fake speech generated by CoSG systems remains limited and largely unexplored. In this paper, we introduce CodecFake-Omni, a large-scale dataset specifically designed to advance the study of neural codec-based deepfake speech (CodecFake) detection and promote progress within the anti-spoofing community. To the best of our knowledge, CodecFake-Omni is the largest dataset of its kind till writing this paper, encompassing the most diverse range of codec architectures. The training set is generated through re-synthesis using nearly all publicly available open-source 31 neural audio codec models across 21 different codec families (one codec family with different configurations will result in multiple different codec models). The evaluation set includes web-sourced data collected from websites generated by 17 advanced CoSG models with eight codec families. Using this large-scale dataset, we reaffirm our previous findings that anti-spoofing models trained on traditional spoofing datasets generated by vocoders struggle to detect synthesized speech from current CoSG systems. Additionally, we propose a comprehensive neural audio codec taxonomy, categorizing neural audio codecs by their root components: vector quantizer, auxiliary objectives, and decoder types, with detailed explanations and representative examples for each. Using this comprehensive taxonomy, we conduct stratified analysis to provide valuable insights for future CodecFake detection research., Comment: Work in Progress: The first two authors contributed equally to this work. Their names are listed alphabetically by first name

Doping an antiferromagnetic Mott insulator is central to our understanding of a variety of phenomena in strongly-correlated electrons, including high-temperature superconductors. To describe the competition between tunneling $t$ of hole dopants and antiferromagnetic (AFM) spin interactions $J$, theoretical and numerical studies often focus on the paradigmatic $t$-$J$ model, and the direct analog quantum simulation of this model in the relevant regime of high-particle density has long been sought. Here, we realize a doped quantum antiferromagnet with next-nearest neighbour (NNN) tunnelings $t'$ and hard-core bosonic holes using a Rydberg tweezer platform. We utilize coherent dynamics between three Rydberg levels, encoding spins and holes, to implement a tunable bosonic $t$-$J$-$V$ model allowing us to study previously inaccessible parameter regimes. We observe dynamical phase separation between hole and spin domains for $|t/J|\ll 1$, and demonstrate the formation of repulsively bound hole pairs in a variety of spin backgrounds. The interference between NNN tunnelings $t'$ and perturbative pair tunneling gives rise to light and heavy pairs depending on the sign of $t$. Using the single-site control allows us to study the dynamics of a single hole in 2D square lattice (anti)ferromagnets. The model we implement extends the toolbox of Rydberg tweezer experiments beyond spin-1/2 models to a larger class of $t$-$J$ and spin-$1$ models., Comment: 8 pages, 5 figures

A Vec-variety is a suitable functor from finite-dimensional vector spaces to finite-dimensional varieties. Most varieties in the geometry of tensors, e.g. the variety of d-way tensors of slice rank at most r, are of this form. We prove that the singular locus of a Vec-variety is a proper closed Vec-subvariety, analogously to the situation for ordinary finite-dimensional varieties. Via earlier work of the third author, this implies that these singular loci admit a description by finitely many polynomial equations. A natural follow-up question to our main result is whether a Vec-variety also admits a suitably functorial resolution of singularities. We establish some preliminary results in this direction in the regime where the dimension of evaluations of a Vec-variety grows linearly with that of the input vector space.

The development of vision-language models (VLMs) is driven by large-scale and diverse multimodal datasets. However, progress toward generalist biomedical VLMs is limited by the lack of annotated, publicly accessible datasets across biology and medicine. Existing efforts are restricted to narrow domains, missing the full diversity of biomedical knowledge encoded in scientific literature. To address this gap, we introduce BIOMEDICA, a scalable, open-source framework to extract, annotate, and serialize the entirety of the PubMed Central Open Access subset into an easy-to-use, publicly accessible dataset. Our framework produces a comprehensive archive with over 24 million unique image-text pairs from over 6 million articles. Metadata and expert-guided annotations are also provided. We demonstrate the utility and accessibility of our resource by releasing BMCA-CLIP, a suite of CLIP-style models continuously pre-trained on the BIOMEDICA dataset via streaming, eliminating the need to download 27 TB of data locally. On average, our models achieve state-of-the-art performance across 40 tasks - spanning pathology, radiology, ophthalmology, dermatology, surgery, molecular biology, parasitology, and cell biology - excelling in zero-shot classification with a 6.56% average improvement (as high as 29.8% and 17.5% in dermatology and ophthalmology, respectively), and stronger image-text retrieval, all while using 10x less compute. To foster reproducibility and collaboration, we release our codebase and dataset for the broader research community.

Deep learning models trained on finite data lack a complete understanding of the physical world. On the other hand, physics-informed neural networks (PINNs) are infused with such knowledge through the incorporation of mathematically expressible laws of nature into their training loss function. By complying with physical laws, PINNs provide advantages over purely data-driven models in limited-data regimes. This feature has propelled them to the forefront of scientific machine learning, a domain characterized by scarce and costly data. However, the vision of accurate physics-informed learning comes with significant challenges. This review examines PINNs for the first time in terms of model optimization and generalization, shedding light on the need for new algorithmic advances to overcome issues pertaining to the training speed, precision, and generalizability of today's PINN models. Of particular interest are the gradient-free methods of neuroevolution for optimizing the uniquely complex loss landscapes arising in PINN training. Methods synergizing gradient descent and neuroevolution for discovering bespoke neural architectures and balancing multiple conflicting terms in physics-informed learning objectives are positioned as important avenues for future research. Yet another exciting track is to cast neuroevolution as a meta-learner of generalizable PINN models., Comment: 20 pages, 8 figures, 1 table

This study explores speaker-specific features encoded in speaker embeddings and intermediate layers of speech self-supervised learning (SSL) models. By utilising a probing method, we analyse features such as pitch, tempo, and energy across prominent speaker embedding models and speech SSL models, including HuBERT, WavLM, and Wav2vec 2.0. The results reveal that speaker embeddings like CAM++ excel in energy classification, while speech SSL models demonstrate superior performance across multiple features due to their hierarchical feature encoding. Intermediate layers effectively capture a mix of acoustic and para-linguistic information, with deeper layers refining these representations. This investigation provides insights into model design and highlights the potential of these representations for downstream applications, such as speaker verification and text-to-speech synthesis, while laying the groundwork for exploring additional features and advanced probing methods.

1. An understanding of how biodiversity confers ecosystem stability is crucial in managing ecosystems under major environmental changes. Multiple biodiversity drivers can stabilize ecosystem functions over time. However, we know little about how local environmental conditions can influence these biodiversity drivers, and consequently how they indirectly shape the ecological stability of ecosystems. 2. We hypothesized that environmental factors can have opposite influences (i.e., not necessarily either positive or negative) on the temporal stability of communities in different environmental ranges depending on the biodiversity drivers involved. We tested this novel hypothesis by using data from a 4-year-long field study of submerged macrophyte across a water depth gradient in 8 heterogeneous bays of Erhai lake (with total sample size of 30,071 quadrats), a large lentic system in China. 3. Results indicate that a unimodal pattern of stability in temporal biomass measurements occurred along the water-depth gradient, and that multiple biodiversity drivers (the asynchrony in species dynamics, and the stability of dominant species) generally increased the temporal stability of aquatic primary producers. However, the effect of water depth either increased or decreased the stability of biomass according to the environmental conditions associated with sites along the water depth gradient. 4. Synthesis. These results reveal the influence of local environmental conditions on the biodiversity drivers of stability may help predict the functional consequences of biodiversity change across different scenarios of environmental change.

The traditional interpretation of Intravascular Ultrasound (IVUS) images during Percutaneous Coronary Intervention (PCI) is time-intensive and inconsistent, relying heavily on physician expertise. Regulatory restrictions and privacy concerns further hinder data integration across hospital systems, complicating collaborative analysis. To address these challenges, a parallel 2D U-Net model with a multi-stage segmentation architecture has been developed, utilizing federated learning to enable secure data analysis across institutions while preserving privacy. The model segments plaques by identifying and subtracting the External Elastic Membrane (EEM) and lumen areas, with preprocessing converting Cartesian to polar coordinates for improved computational efficiency. Achieving a Dice Similarity Coefficient (DSC) of 0.706, the model effectively identifies plaques and detects circular boundaries in real-time. Collaborative efforts with domain experts enhance plaque burden interpretation through precise quantitative measurements. Future advancements may involve integrating advanced federated learning techniques and expanding datasets to further improve performance and applicability. This adaptable technology holds promise for environments handling sensitive, distributed data, offering potential to optimize outcomes in medical imaging and intervention.

Symmetry plays a central role in the sciences, machine learning, and statistics. While statistical tests for the presence of distributional invariance with respect to groups have a long history, tests for conditional symmetry in the form of equivariance or conditional invariance are absent from the literature. This work initiates the study of nonparametric randomization tests for symmetry (invariance or equivariance) of a conditional distribution under the action of a specified locally compact group. We develop a general framework for randomization tests with finite-sample Type I error control and, using kernel methods, implement tests with finite-sample power lower bounds. We also describe and implement approximate versions of the tests, which are asymptotically consistent. We study their properties empirically on synthetic examples, and on applications to testing for symmetry in two problems from high-energy particle physics., Comment: Theorems 2.2 and 4.1 appeared in arXiv:2307.15834, which is superseded by this article

Orthognathic surgery consultation is essential to help patients understand the changes to their facial appearance after surgery. However, current visualization methods are often inefficient and inaccurate due to limited pre- and post-treatment data and the complexity of the treatment. To overcome these challenges, this study aims to develop a fully automated pipeline that generates accurate and efficient 3D previews of postsurgical facial appearances for patients with orthognathic treatment without requiring additional medical images. The study introduces novel aesthetic losses, such as mouth-convexity and asymmetry losses, to improve the accuracy of facial surgery prediction. Additionally, it proposes a specialized parametric model for 3D reconstruction of the patient, medical-related losses to guide latent code prediction network optimization, and a data augmentation scheme to address insufficient data. The study additionally employs FLAME, a parametric model, to enhance the quality of facial appearance previews by extracting facial latent codes and establishing dense correspondences between pre- and post-surgery geometries. Quantitative comparisons showed the algorithm's effectiveness, and qualitative results highlighted accurate facial contour and detail predictions. A user study confirmed that doctors and the public could not distinguish between machine learning predictions and actual postoperative results. This study aims to offer a practical, effective solution for orthognathic surgery consultations, benefiting doctors and patients., Comment: 9 pages, 5 figures

Multimodal incremental learning needs to digest the information from multiple modalities while concurrently learning new knowledge without forgetting the previously learned information. There are numerous challenges for this task, mainly including the larger storage size of multimodal data in exemplar-based methods and the computational requirement of finetuning on huge multimodal models. In this paper, we leverage the parameter-efficient tuning scheme to reduce the burden of fine-tuning and propose the exemplar masking framework to efficiently replay old knowledge. Specifically, the non-important tokens are masked based on the attention weights and the correlation across different modalities, significantly reducing the storage size of an exemplar and consequently saving more exemplars under the same memory buffer. Moreover, we design a multimodal data augmentation technique to diversify exemplars for replaying prior knowledge. In experiments, we not only evaluate our method in existing multimodal datasets but also extend the ImageNet-R dataset to a multimodal dataset as a real-world application, where captions are generated by querying multimodal large language models (e.g., InstructBLIP). Extensive experiments show that our exemplar masking framework is more efficient and robust to catastrophic forgetting under the same limited memory buffer. Code is available at https://github.com/YiLunLee/Exemplar_Masking_MCIL., Comment: Project page: https://github.com/YiLunLee/Exemplar_Masking_MCIL

Three-dimensional (3D) understanding of objects and scenes play a key role in humans' ability to interact with the world and has been an active area of research in computer vision, graphics, and robotics. Large scale synthetic and object-centric 3D datasets have shown to be effective in training models that have 3D understanding of objects. However, applying a similar approach to real-world objects and scenes is difficult due to a lack of large-scale data. Videos are a potential source for real-world 3D data, but finding diverse yet corresponding views of the same content has shown to be difficult at scale. Furthermore, standard videos come with fixed viewpoints, determined at the time of capture. This restricts the ability to access scenes from a variety of more diverse and potentially useful perspectives. We argue that large scale 360 videos can address these limitations to provide: scalable corresponding frames from diverse views. In this paper, we introduce 360-1M, a 360 video dataset, and a process for efficiently finding corresponding frames from diverse viewpoints at scale. We train our diffusion-based model, Odin, on 360-1M. Empowered by the largest real-world, multi-view dataset to date, Odin is able to freely generate novel views of real-world scenes. Unlike previous methods, Odin can move the camera through the environment, enabling the model to infer the geometry and layout of the scene. Additionally, we show improved performance on standard novel view synthesis and 3D reconstruction benchmarks., Comment: NeurIPS 2024. For project page, see https://mattwallingford.github.io/ODIN

While large vision-language models (LVLMs) have shown impressive capabilities in generating plausible responses correlated with input visual contents, they still suffer from hallucinations, where the generated text inaccurately reflects visual contents. To address this, recent approaches apply contrastive decoding to calibrate the model's response via contrasting output distributions with original and visually distorted samples, demonstrating promising hallucination mitigation in a training-free manner. However, the potential of changing information in visual inputs is not well-explored, so a deeper investigation into the behaviors of visual contrastive decoding is of great interest. In this paper, we first explore various methods for contrastive decoding to change visual contents, including image downsampling and editing. Downsampling images reduces the detailed textual information while editing yields new contents in images, providing new aspects as visual contrastive samples. To further study benefits by using different contrastive samples, we analyze probability-level metrics, including entropy and distribution distance. Interestingly, the effect of these samples in mitigating hallucinations varies a lot across LVLMs and benchmarks. Based on our analysis, we propose a simple yet effective method to combine contrastive samples, offering a practical solution for applying contrastive decoding across various scenarios. Extensive experiments are conducted to validate the proposed fusion method among different benchmarks., Comment: Under review. Project pages: https://github.com/YiLunLee/VCD_Analysis

Second-order topological superconductors host Majorana corner modes (MCMs), which are confined to specific corners of the system. This spatial restriction presents challenges for manipulating and relocating MCMs. We propose a novel protocol for dynamically controlling the movement of time-reversal symmetric MCMs in a heterostructure consisting of a 2D Weyl semimetal and a $d$-wave superconductor. By leveraging the energy asymmetry of topological edge states in the 2D Weyl semimetal, the position of MCMs can be effectively tuned via chemical potential adjustments. We further introduce a device architecture that integrates multiple heterostructure blocks, each controlled by independent gate voltages, to enable the adiabatic movement and exchange of MCMs. This approach demonstrates a robust mechanism for Majorana manipulation and provides a scalable framework for future experimental studies of topological quantum computation., Comment: 5 figures

We use the determinant method of Bombieri-Pila and Heath-Brown and its Arakelov reformulation by Chen utilizing Bost's slope method to estimate the number of hypersurfaces required to cover the regular rational points with bounded Arakelov height on a projective variety. Using complex cellular structures introduced by Binyamini-Novikov, we replace the usual subpolynomial factor by a polylogarithmic factor in the estimation., Comment: 13 pages, comments welcome!

Electrocardiogram (ECG) signals play a crucial role in diagnosing cardiovascular diseases. To reduce power consumption in wearable or portable devices used for long-term ECG monitoring, super-resolution (SR) techniques have been developed, enabling these devices to collect and transmit signals at a lower sampling rate. In this study, we propose MSECG, a compact neural network model designed for ECG SR. MSECG combines the strength of the recurrent Mamba model with convolutional layers to capture both local and global dependencies in ECG waveforms, allowing for the effective reconstruction of high-resolution signals. We also assess the model's performance in real-world noisy conditions by utilizing ECG data from the PTB-XL database and noise data from the MIT-BIH Noise Stress Test Database. Experimental results show that MSECG outperforms two contemporary ECG SR models under both clean and noisy conditions while using fewer parameters, offering a more powerful and robust solution for long-term ECG monitoring applications., Comment: 5 pages, 3 figures

We propose an efficient radiance field rendering algorithm that incorporates a rasterization process on sparse voxels without neural networks or 3D Gaussians. There are two key contributions coupled with the proposed system. The first is to render sparse voxels in the correct depth order along pixel rays by using dynamic Morton ordering. This avoids the well-known popping artifact found in Gaussian splatting. Second, we adaptively fit sparse voxels to different levels of detail within scenes, faithfully reproducing scene details while achieving high rendering frame rates. Our method improves the previous neural-free voxel grid representation by over 4db PSNR and more than 10x rendering FPS speedup, achieving state-of-the-art comparable novel-view synthesis results. Additionally, our neural-free sparse voxels are seamlessly compatible with grid-based 3D processing algorithms. We achieve promising mesh reconstruction accuracy by integrating TSDF-Fusion and Marching Cubes into our sparse grid system., Comment: Code release in progress

Virtual models of human gait, or digital twins, offer a promising solution for studying mobility without the need for labor-intensive data collection. However, challenges such as the sim-to-real gap and limited adaptability to diverse walking conditions persist. To address these, we developed and validated a framework to create a skeletal humanoid agent capable of adapting to varying walking speeds while maintaining biomechanically realistic motions. The framework combines a synthetic data generator, which produces biomechanically plausible gait kinematics from open-source biomechanics data, and a training system that uses adversarial imitation learning to train the agent's walking policy. We conducted comprehensive analyses comparing the agent's kinematics, synthetic data, and the original biomechanics dataset. The agent achieved a root mean square error of 5.24 +- 0.09 degrees at varying speeds compared to ground-truth kinematics data, demonstrating its adaptability. This work represents a significant step toward developing a digital twin of human locomotion, with potential applications in biomechanics research, exoskeleton design, and rehabilitation., Comment: Currently under review

In vision-language models (VLMs), the ability to perceive and interpret color and physical environment is crucial for achieving contextually accurate understanding and interaction. However, despite advances in multimodal modeling, there remains a significant lack of specialized datasets that rigorously evaluate a model's capacity to discern subtle color variations and spatial context -- critical elements for situational comprehension and reliable deployment across real-world applications. Toward that goal, we curate MegaCOIN, a high-quality, human-labeled dataset based on \emph{real} images with various contextual attributes. MegaCOIN consists of two parts: MegaCOIN-Instruct, which serves as a supervised fine-tuning (SFT) dataset for VLMs; and MegaCOIN-Bench, an annotated test set that can be used as a stand-alone QA dataset. MegaCOIN~provides three annotated features for 220,000 real images: foreground color, background color, and description of an object's physical environment, constituting 660k human annotations. In addition, MegaCOIN can be applied to benchmark domain generalization (DG) algorithms. We explore benchmarking DG methods in the linear probing setup for VLM and show some new insights. Last but not least, we show that VLMs, including GPT-4o, have subpar color recognition capabilities, and fine-tuning with MegaCOIN can result in improved performance on visual evaluation tasks. In certain cases, MegaCOIN fine-tuned small-scale opensource models such as LLaVA and Bunny can outperform closed-source GPT-4o. We hope the utilities of MegaCOIN can shed light on the directions VLMs can improve and provide a more complex platform for domain generalization algorithms., Comment: 8 pages, 13 tables, 2 figures

We introduce FathomGPT, an open source system for the interactive investigation of ocean science data via a natural language interface. FathomGPT was developed in close collaboration with marine scientists to enable researchers to explore and analyze the FathomNet image database. FathomGPT provides a custom information retrieval pipeline that leverages OpenAI's large language models to enable: the creation of complex queries to retrieve images, taxonomic information, and scientific measurements; mapping common names and morphological features to scientific names; generating interactive charts on demand; and searching by image or specified patterns within an image. In designing FathomGPT, particular emphasis was placed on enhancing the user's experience by facilitating free-form exploration and optimizing response times. We present an architectural overview and implementation details of FathomGPT, along with a series of ablation studies that demonstrate the effectiveness of our approach to name resolution, fine tuning, and prompt modification. We also present usage scenarios of interactive data exploration sessions and document feedback from ocean scientists and machine learning experts., Comment: The first two authors contributed equally to this work. Accepted to the 37th Annual ACM Symposium on User Interface Software and Technology (UIST 2024)

With the goal of benchmarking generative systems beyond expert software development ability, we introduce Commit0, a benchmark that challenges AI agents to write libraries from scratch. Agents are provided with a specification document outlining the library's API as well as a suite of interactive unit tests, with the goal of producing an implementation of this API accordingly. The implementation is validated through running these unit tests. As a benchmark, Commit0 is designed to move beyond static one-shot code generation towards agents that must process long-form natural language specifications, adapt to multi-stage feedback, and generate code with complex dependencies. Commit0 also offers an interactive environment where models receive static analysis and execution feedback on the code they generate. Our experiments demonstrate that while current agents can pass some unit tests, none can yet fully reproduce full libraries. Results also show that interactive feedback is quite useful for models to generate code that passes more unit tests, validating the benchmarks that facilitate its use.