Search

Video compression plays a pivotal role in managing and transmitting large-scale display data, particularly given the growing demand for higher resolutions and improved video quality. This paper proposes an optimized memory system architecture for Video Electronics Standards Association (VESA) Display Compression-M (VDC-M) decoder, characterized by its substantial on-chip buffer requirements. We design and analyze three architectures categorized by optimization levels and management complexity. Our strategy focuses on enhancing line buffer access scheduling and minimizing reconstruction buffer, targeting prediction and multi-slice operation that are the major resource consumers in the decoder. By adjusting line delay and segmenting SRAM bank alongside reconstructed block forwarding, we achieve a 33.3% size reduction in the line buffer and 77.3% in the reconstruction buffer compared to Baseline VDC-M decoder. Synthesized using a 28 nm CMOS process, the proposed architecture achieves a 31.5% reduction in gate count of the decoder backend hardware, supporting real-time performance with up to 96.45 fps for 4K UHD resolution at 200 MHz operating frequency and a throughput of 4 pixels per cycle., Comment: Accepted by 2025 IEEE International Symposium on Circuits and Systems (ISCAS)

The circumstellar envelope of the carbon star CW Leo exhibited various unexpected changes in recent optical imaging observations. We have performed a follow-up observation using the Near-infrared Integral-Field Spectrograph (NIFS) equipped on the Gemini-North telescope. We report the near-infrared counterparts of a local brightness peak in the optical at the stellar position of CW Leo. On the other hand, a second peak detected at short wavelengths in the J band coincides with the brightest, bluest position in the optical images. The absorption features in the K band are minimized at a radius of 0.2 arcsec from the predicted stellar position. The reduction of the absorption depths likely indicates dilution of the absorption features by thermal emission of dust grains newly formed at such a radius and heated by radiation from the central star. The broad absorption feature at 1.53 um is significantly deeper than in template carbon stars, consistent with the presence of a substantial amount of circumstellar material around CW Leo. Its northeastern quadrant lacks circumstellar absorption features and scattered light in the near-infrared regime, which are possibly manifestations of its conical cavity in both gas and dust. In addition, a cross correlation of CO overtone bands indicates that the average expansion velocity of dust grains is smaller to the northern direction, likewise the velocity of transverse wind components derived using the differential proper motion of a circumstellar whirled pattern. The gradual brightening of CW Leo and the changes in its innermost circumstellar envelope need further continuous monitoring observations to properly understand its transitional phase toward the post-asymptotic-giant-branch stage., Comment: 18 pages, 13 figures, 1 table, Published in AJ

This paper examines how gifting spreads among viewers on Twitch, one of the largest live streaming platforms worldwide. Twitch users can give gift subscriptions to other viewers in the chat room, with the majority of gifters opting for community gifting, which is gifting to randomly selected viewers. We identify the random nature of gift-receiving in our data as a natural experiment setting. We investigate whether gift recipients pay it forward, considering various gift types that may either promote or deter the spread of gifting. Our findings reveal that Twitch viewers who receive gift subscriptions are generally more likely to pay it forward than non-recipients, and the positive impact of gift-receiving becomes stronger when the recipient is the sole beneficiary of the giver's gifting behavior. However, we found that gifts from frequent gifters discourage recipients from paying it forward, and gifts from anonymous gifters do not influence the likelihood of viewers becoming future gifters. This research contributes to the existing literature on the spread of online prosocial behavior by providing robust evidence and suggests practical strategies for promoting online gifting.

Over the past few years, diffusion models have emerged as novel AI solutions, generating diverse multi-modal outputs from text prompts. Despite their capabilities, they face challenges in computing, such as excessive latency and energy consumption due to their iterative architecture. Although prior works specialized in transformer acceleration can be applied, the iterative nature of diffusion models remains unresolved. In this paper, we present EXION, the first SW-HW co-designed diffusion accelerator that solves the computation challenges by exploiting the unique inter- and intra-iteration output sparsity in diffusion models. To this end, we propose two SW-level optimizations. First, we introduce the FFN-Reuse algorithm that identifies and skips redundant computations in FFN layers across different iterations (inter-iteration sparsity). Second, we use a modified eager prediction method that employs two-step leading-one detection to accurately predict the attention score, skipping unnecessary computations within an iteration (intra-iteration sparsity). We also introduce a novel data compaction mechanism named ConMerge, which can enhance HW utilization by condensing and merging sparse matrices into compact forms. Finally, it has a dedicated HW architecture that supports the above sparsity-inducing algorithms, translating high output sparsity into improved energy efficiency and performance. To verify the feasibility of the EXION, we first demonstrate that it has no impact on accuracy in various types of multi-modal diffusion models. We then instantiate EXION in both server- and edge-level settings and compare its performance against GPUs with similar specifications. Our evaluation shows that EXION achieves dramatic improvements in performance and energy efficiency by 3.2-379.3x and 45.1-3067.6x compared to a server GPU and by 42.6-1090.9x and 196.9-4668.2x compared to an edge GPU., Comment: To appear in 2025 IEEE International Symposium on High-Performance Computer Architecture (HPCA 2025)

The goal of voice conversion is to transform the speech of a source speaker to sound like that of a reference speaker while preserving the original content. A key challenge is to extract disentangled linguistic content from the source and voice style from the reference. While existing approaches leverage various methods to isolate the two, a generalization still requires further attention, especially for robustness in zero-shot scenarios. In this paper, we achieve successful disentanglement of content and speaker features by tuning self-supervised speech features with adapters. The adapters are trained to dynamically encode nuanced features from rich self-supervised features, and the decoder fuses them to produce speech that accurately resembles the reference with minimal loss of content. Moreover, we leverage a conditional flow matching decoder with cross-attention speaker conditioning to further boost the synthesis quality and efficiency. Subjective and objective evaluations in a zero-shot scenario demonstrate that the proposed method outperforms existing models in speech quality and similarity to the reference speech., Comment: ICASSP 2025; demo available https://mm.kaist.ac.kr/projects/AdaptVC

The goal of this work is to generate natural speech in multiple languages while maintaining the same speaker identity, a task known as cross-lingual speech synthesis. A key challenge of cross-lingual speech synthesis is the language-speaker entanglement problem, which causes the quality of cross-lingual systems to lag behind that of intra-lingual systems. In this paper, we propose CrossSpeech++, which effectively disentangles language and speaker information and significantly improves the quality of cross-lingual speech synthesis. To this end, we break the complex speech generation pipeline into two simple components: language-dependent and speaker-dependent generators. The language-dependent generator produces linguistic variations that are not biased by specific speaker attributes. The speaker-dependent generator models acoustic variations that characterize speaker identity. By handling each type of information in separate modules, our method can effectively disentangle language and speaker representation. We conduct extensive experiments using various metrics, and demonstrate that CrossSpeech++ achieves significant improvements in cross-lingual speech synthesis, outperforming existing methods by a large margin.

Using the Subaru/FOCAS IFU capability, we examine the spatially resolved relationships between gas-phase metallicity, stellar mass, and star-formation rate surface densities (Sigma_* and Sigma_SFR, respectively) in extremely metal-poor galaxies (EMPGs) in the local universe. Our analysis includes 24 EMPGs, comprising 9,177 spaxels, which span a unique parameter space of local metallicity (12+log(O/H) = 6.9 to 7.9) and stellar mass surface density (Sigma_* ~ 10^5 to 10^7 Msun/kpc^2), extending beyond the range of existing large integral-field spectroscopic surveys. Through spatially resolved emission line diagnostics based on the [NII] BPT-diagram, we verify the absence of evolved active galactic nuclei in these EMPGs. Our findings reveal that, while the resolved mass-metallicity relation exhibits significant scatter in the low-mass regime, this scatter is closely correlated with local star-formation surface density. Specifically, metallicity decreases as Sigma_SFR increases for a given Sigma_*. Notably, half of the EMPGs show a distinct metal-poor horizontal branch on the resolved mass-metallicity relation. This feature typically appears at the peak clump with the highest Sigma_* and Sigma_SFR and is surrounded by a relatively metal-enriched ambient region. These findings support a scenario in which metal-poor gas infall fuels episodic star formation in EMPGs, consistent with the kinematic properties observed in these systems. In addition, we identify four EMPGs with exceptionally low central metallicities (12+log(O/H) <~ 7.2), which display only a metal-poor clump without a surrounding metal-rich region. This suggests that such ultra-low metallicity EMPGs, at less than a few percent of the solar metallicity, may serve as valuable analogs for galaxies in the early stages of galaxy evolution., Comment: 29 pages, 14 figures. Submitted to ApJ

In this paper, we introduce V2SFlow, a novel Video-to-Speech (V2S) framework designed to generate natural and intelligible speech directly from silent talking face videos. While recent V2S systems have shown promising results on constrained datasets with limited speakers and vocabularies, their performance often degrades on real-world, unconstrained datasets due to the inherent variability and complexity of speech signals. To address these challenges, we decompose the speech signal into manageable subspaces (content, pitch, and speaker information), each representing distinct speech attributes, and predict them directly from the visual input. To generate coherent and realistic speech from these predicted attributes, we employ a rectified flow matching decoder built on a Transformer architecture, which models efficient probabilistic pathways from random noise to the target speech distribution. Extensive experiments demonstrate that V2SFlow significantly outperforms state-of-the-art methods, even surpassing the naturalness of ground truth utterances.

Extraction of hot carriers (HCs) over the band-edge is a key to harvest solar energy beyond Shockley-Queisser limit1. Graphene is known as a HC-layered material due to phonon bottleneck effect near Dirac point, but limited by low photocarrier density2. Graphene/transition metal dichalcogenide (TMD) heterostructures circumvent this issue by ultrafast carrier transfer from TMD to graphene2,3. Nevertheless, efficient extraction of photocurrent by means of HCs together with carrier multiplication (CM) is still missing. Here, we introduce an ultrathin broadband resonant tunneling (BRT) barrier, TiOX to efficiently extract photocurrent with simultaneous CM and HC measurements in MoS2/graphene/TiOX heterostructure. The BRT layer gives rise to boosting open circuit voltage which is linearly proportional to incident photon energy. Meanwhile, short circuit current rises rapidly over 2Eg with obvious CM feature. This was explained by defining the joint density of states between graphene and TiOX layer over positive and negative voltage. The broadband resonant tunneling states inherently constructed from oxidation states varying from Ti3+ to Ti4+ allow the ultrafast HCs to efficiently transfer from graphene to TiOX layer. We find that the number of available tunneling states is directly proportional to short circuit current, which is well corroborated with TiOX and MoS2 thickness variance. We obtained an optimum thickness of BRT layer of ~2.8 nm, yielding cascade open circuit voltage as high as ~0.7 V, two orders of magnitude higher than that without BRT layer to reach a record efficiency of 5.3% with improved fill factor owing to synergistic HC and CM conversion under 1-SUN with long-term stability.

Surgical phase recognition (SPR) is an integral component of surgical data science, enabling high-level surgical analysis. End-to-end trained neural networks that predict surgical phase directly from videos have shown excellent performance on benchmarks. However, these models struggle with robustness due to non-causal associations in the training set. Our goal is to improve model robustness to variations in the surgical videos by leveraging the digital twin (DT) paradigm -- an intermediary layer to separate high-level analysis (SPR) from low-level processing. As a proof of concept, we present a DT representation-based framework for SPR from videos. The framework employs vision foundation models with reliable low-level scene understanding to craft DT representation. We embed the DT representation in place of raw video inputs in the state-of-the-art SPR model. The framework is trained on the Cholec80 dataset and evaluated on out-of-distribution (OOD) and corrupted test samples. Contrary to the vulnerability of the baseline model, our framework demonstrates strong robustness on both OOD and corrupted samples, with a video-level accuracy of 80.3 on a highly corrupted Cholec80 test set, 67.9 on the challenging CRCD dataset, and 99.8 on an internal robotic surgery dataset, outperforming the baseline by 3.9, 16.8, and 90.9 respectively. We also find that using DT representation as an augmentation to the raw input can significantly improve model robustness. Our findings lend support to the thesis that DT representations are effective in enhancing model robustness. Future work will seek to improve the feature informativeness and incorporate interpretability for a more comprehensive framework.

The goal of this paper is to accelerate codec-based speech synthesis systems with minimum sacrifice to speech quality. We propose an enhanced inference method that allows for flexible trade-offs between speed and quality during inference without requiring additional training. Our core idea is to predict multiple tokens per inference step of the AR module using multiple prediction heads, resulting in a linear reduction in synthesis time as the number of heads increases. Furthermore, we introduce a novel speculative decoding technique that utilises a Viterbi-based algorithm to select the optimal sequence of generated tokens at each decoding step. In our experiments, we demonstrate that the time required to predict each token is reduced by a factor of 4 to 5 compared to baseline models, with minimal quality trade-off or even improvement in terms of speech intelligibility. Audio samples are available at: multpletokensprediction.github.io/multipletokensprediction.github.io/., Comment: Submitted to IEEE ICASSP 2025

This paper introduces SpoofCeleb, a dataset designed for Speech Deepfake Detection (SDD) and Spoofing-robust Automatic Speaker Verification (SASV), utilizing source data from real-world conditions and spoofing attacks generated by Text-To-Speech (TTS) systems also trained on the same real-world data. Robust recognition systems require speech data recorded in varied acoustic environments with different levels of noise to be trained. However, existing datasets typically include clean, high-quality recordings (bona fide data) due to the requirements for TTS training; studio-quality or well-recorded read speech is typically necessary to train TTS models. Existing SDD datasets also have limited usefulness for training SASV models due to insufficient speaker diversity. We present SpoofCeleb, which leverages a fully automated pipeline that processes the VoxCeleb1 dataset, transforming it into a suitable form for TTS training. We subsequently train 23 contemporary TTS systems. The resulting SpoofCeleb dataset comprises over 2.5 million utterances from 1,251 unique speakers, collected under natural, real-world conditions. The dataset includes carefully partitioned training, validation, and evaluation sets with well-controlled experimental protocols. We provide baseline results for both SDD and SASV tasks. All data, protocols, and baselines are publicly available at https://jungjee.github.io/spoofceleb., Comment: 9 pages, 2 figures, 8 tables

Text-to-speech (TTS) systems are traditionally trained using modest databases of studio-quality, prompted or read speech collected in benign acoustic environments such as anechoic rooms. The recent literature nonetheless shows efforts to train TTS systems using data collected in the wild. While this approach allows for the use of massive quantities of natural speech, until now, there are no common datasets. We introduce the TTS In the Wild (TITW) dataset, the result of a fully automated pipeline, in this case, applied to the VoxCeleb1 dataset commonly used for speaker recognition. We further propose two training sets. TITW-Hard is derived from the transcription, segmentation, and selection of VoxCeleb1 source data. TITW-Easy is derived from the additional application of enhancement and additional data selection based on DNSMOS. We show that a number of recent TTS models can be trained successfully using TITW-Easy, but that it remains extremely challenging to produce similar results using TITW-Hard. Both the dataset and protocols are publicly available and support the benchmarking of TTS systems trained using TITW data., Comment: 5 pages, submitted to ICASSP 2025 as a conference paper

We present a proof-of-concept simulation-based inference on $\Omega_{\rm m}$ and $\sigma_{8}$ from the SDSS BOSS LOWZ NGC catalog using neural networks and domain generalization techniques without the need of summary statistics. Using rapid lightcone simulations, ${\rm L{\scriptsize -PICOLA}}$, mock galaxy catalogs are produced that fully incorporate the observational effects. The collection of galaxies is fed as input to a point cloud-based network, ${\texttt{Minkowski-PointNet}}$. We also add relatively more accurate ${\rm G{\scriptsize ADGET}}$ mocks to obtain robust and generalizable neural networks. By explicitly learning the representations which reduces the discrepancies between the two different datasets via the semantic alignment loss term, we show that the latent space configuration aligns into a single plane in which the two cosmological parameters form clear axes. Consequently, during inference, the SDSS BOSS LOWZ NGC catalog maps onto the plane, demonstrating effective generalization and improving prediction accuracy compared to non-generalized models. Results from the ensemble of 25 independently trained machines find $\Omega_{\rm m}=0.339 \pm 0.056$ and $\sigma_{8}=0.801 \pm 0.061$, inferred only from the distribution of galaxies in the lightcone slices without relying on any indirect summary statistics. A single machine that best adapts to the ${\rm G{\scriptsize ADGET}}$ mocks yields a tighter prediction of $\Omega_{\rm m}=0.282 \pm 0.014$ and $\sigma_{8}=0.786 \pm 0.036$. We emphasize that adaptation across multiple domains can enhance the robustness of the neural networks in observational data., Comment: 28 pages, 14 figures; Accepted for publication in ApJ

The mass-radius relationship of white dwarfs (WDs) is one of their defining characteristics, largely derived from electron degeneracy pressure. We present a model-independent study of the observed mass-radius relationship in WD binaries of \cite{Parsons_2017}, listing data over a broad temperature range up to about 60,000 K (5 eV). The data show an appreciable temperature sensitivity with pronounced intrinsic scatter (beyond measurement uncertainty) for the canonical He-models with proton-to-neutron ratio 1:1. We characterize temperature sensitivity by a temperature scale $T_0$ in model-agnostic power-law relations with temperature normalized radius. For low-mass WDs, the results identify a remarkably modest $T_0 = 1 \sim 2 $ eV. We comment on a potential interpretation for atmospheres insulating super-Eddington temperature cores from the sub-Eddington photospheres of low-mass WDs., Comment: 8 pages, 6 figures, to appear in NewA

We introduce a GPU-accelerated hybrid hydro/N-body code (Enzo-N) designed to address the challenges of concurrently simulating star clusters and their parent galaxies. This task has been exceedingly challenging, primarily due to the considerable computational time required, which stems from the substantial scale difference between galaxies (~ 0.1 Mpc) and star clusters (~ pc). Yet, this significant scale separation means that particles within star clusters perceive those outside the star cluster in a semi-stationary state. By leveraging this aspect, we integrate the direct N-body code (Nbody6++GPU) into the cosmological (magneto-)hydrodynamic code (Enzo) through the utilization of the semi-stationary background acceleration approximation. We solve the dynamics of particles within star clusters using the direct N-body solver with regularization for few-body interactions, while evolving particles outside -- dark matter, gas, and stars -- using the particle-mesh gravity solver and hydrodynamic methods. We demonstrate that Enzo-N successfully simulates the co-evolution of star clusters and their parent galaxies, capturing phenomena such as core collapse of the star cluster and tidal stripping due to galactic tides. This comprehensive framework opens up new possibilities for studying the evolution of star clusters within galaxies, offering insights that were previously inaccessible., Comment: 15 pages, 13 figures, Accepted for publication in ApJ

The AGORA Cosmorun (arXiv:2106.09738) is a set of hydrodynamical cosmological zoom-in simulations carried out within the AGORA High-resolution Galaxy Simulations Comparison Project (arXiv:1308.2669,arXiv:1610.03066). These simulations show the formation and evolution of a Milky Way-sized galaxy using eight of the most widely used numerical codes in the community (Art-I, Enzo, Ramses, Changa, Gadget-3, Gear, Gizmo, and Arepo). In this short report, we describe the public release of the raw output data from all of these simulations at z = 8, 7, 6, 5, 4, 3, 2 (plus at z=1, 0 when available), and several metadata files containing the halo centers, virial quantities, and merger trees. The data from even thinner timesteps will be released as soon as the upcoming collaboration papers (VII-IX) are submitted and accepted., Comment: ArXiv Data Release from the AGORA Collaboration. arXiv admin note: text overlap with arXiv:2001.04354

This paper introduces VoxSim, a dataset of perceptual voice similarity ratings. Recent efforts to automate the assessment of speech synthesis technologies have primarily focused on predicting mean opinion score of naturalness, leaving speaker voice similarity relatively unexplored due to a lack of extensive training data. To address this, we generate about 41k utterance pairs from the VoxCeleb dataset, a widely utilised speech dataset for speaker recognition, and collect nearly 70k speaker similarity scores through a listening test. VoxSim offers a valuable resource for the development and benchmarking of speaker similarity prediction models. We provide baseline results of speaker similarity prediction models on the VoxSim test set and further demonstrate that the model trained on our dataset generalises to the out-of-domain VCC2018 dataset., Comment: INTERSPEECH 2024. The dataset is available from https://mm.kaist.ac.kr/projects/voxsim/

We introduce a novel software called TCSpy which is designed to efficiently control a multi-telescope array through network-based protocols. The primary objectives of TCSpy include centralized control of the array, support for diverse observation modes, and swift responses to the follow-up observations of astronomical transients. To achieve these objectives, TCSpy utilizes the ASCOM Alpaca protocol in conjunction with Alpyca, establishing robust communication among multiple telescope units. For the practical application of TCSpy, we implement TCSpy within the 7-Dimensional Telescope (7DT). 7DT is a telescope array consisting of 20, 0.5-m telescopes, equipped with 40 different medium-band filters. The main scientific goals of 7DT include detecting the optical counterparts of gravitational-wave sources, identifying kilonovae, and the spectral mapping of the southern sky. Through the integration of TCSpy, 7DT can achieve these scientific objectives with its unique observation modes and rapid follow-up capabilities., Comment: 8 pages, 5 figures, SPIE Astronomical Telescopes + Instrumentation 2024

We explore whether surgical manipulation tasks can be learned on the da Vinci robot via imitation learning. However, the da Vinci system presents unique challenges which hinder straight-forward implementation of imitation learning. Notably, its forward kinematics is inconsistent due to imprecise joint measurements, and naively training a policy using such approximate kinematics data often leads to task failure. To overcome this limitation, we introduce a relative action formulation which enables successful policy training and deployment using its approximate kinematics data. A promising outcome of this approach is that the large repository of clinical data, which contains approximate kinematics, may be directly utilized for robot learning without further corrections. We demonstrate our findings through successful execution of three fundamental surgical tasks, including tissue manipulation, needle handling, and knot-tying., Comment: 8 pages