Search

This technical report presents our initial attempt to build a spoken large language model (LLM) for Taiwanese Mandarin, specifically tailored to enable real-time, speech-to-speech interaction in multi-turn conversations. Our end-to-end model incorporates a decoder-only transformer architecture and aims to achieve seamless interaction while preserving the conversational flow, including full-duplex capabilities allowing simultaneous speaking and listening. The paper also details the training process, including data preparation with synthesized dialogues and adjustments for real-time interaction. We also developed a platform to evaluate conversational fluency and response coherence in multi-turn dialogues. We hope the release of the report can contribute to the future development of spoken LLMs in Taiwanese Mandarin., Comment: Work in progress

Multimodal foundation models, such as Gemini and ChatGPT, have revolutionized human-machine interactions by seamlessly integrating various forms of data. Developing a universal spoken language model that comprehends a wide range of natural language instructions is critical for bridging communication gaps and facilitating more intuitive interactions. However, the absence of a comprehensive evaluation benchmark poses a significant challenge. We present Dynamic-SUPERB Phase-2, an open and evolving benchmark for the comprehensive evaluation of instruction-based universal speech models. Building upon the first generation, this second version incorporates 125 new tasks contributed collaboratively by the global research community, expanding the benchmark to a total of 180 tasks, making it the largest benchmark for speech and audio evaluation. While the first generation of Dynamic-SUPERB was limited to classification tasks, Dynamic-SUPERB Phase-2 broadens its evaluation capabilities by introducing a wide array of novel and diverse tasks, including regression and sequence generation, across speech, music, and environmental audio. Evaluation results indicate that none of the models performed well universally. SALMONN-13B excelled in English ASR, while WavLLM demonstrated high accuracy in emotion recognition, but current models still require further innovations to handle a broader range of tasks. We will soon open-source all task data and the evaluation pipeline.

Time-to-Collision (TTC) estimation lies in the core of the forward collision warning (FCW) functionality, which is key to all Automatic Emergency Braking (AEB) systems. Although the success of solutions using frame-based cameras (e.g., Mobileye's solutions) has been witnessed in normal situations, some extreme cases, such as the sudden variation in the relative speed of leading vehicles and the sudden appearance of pedestrians, still pose significant risks that cannot be handled. This is due to the inherent imaging principles of frame-based cameras, where the time interval between adjacent exposures introduces considerable system latency to AEB. Event cameras, as a novel bio-inspired sensor, offer ultra-high temporal resolution and can asynchronously report brightness changes at the microsecond level. To explore the potential of event cameras in the above-mentioned challenging cases, we propose EvTTC, which is, to the best of our knowledge, the first multi-sensor dataset focusing on TTC tasks under high-relative-speed scenarios. EvTTC consists of data collected using standard cameras and event cameras, covering various potential collision scenarios in daily driving and involving multiple collision objects. Additionally, LiDAR and GNSS/INS measurements are provided for the calculation of ground-truth TTC. Considering the high cost of testing TTC algorithms on full-scale mobile platforms, we also provide a small-scale TTC testbed for experimental validation and data augmentation. All the data and the design of the testbed are open sourced, and they can serve as a benchmark that will facilitate the development of vision-based TTC techniques., Comment: 8 pages, 7 figures, 5 tables

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

3D anomaly detection and localization is of great significance for industrial inspection. Prior 3D anomaly detection and localization methods focus on the setting that the testing data share the same category as the training data which is normal. However, in real-world applications, the normal training data for the target 3D objects can be unavailable due to issues like data privacy or export control regulation. To tackle these challenges, we identify a new task -- zero-shot 3D anomaly detection and localization, where the training and testing classes do not overlap. To this end, we design 3DzAL, a novel patch-level contrastive learning framework based on pseudo anomalies generated using the inductive bias from task-irrelevant 3D xyz data to learn more representative feature representations. Furthermore, we train a normalcy classifier network to classify the normal patches and pseudo anomalies and utilize the classification result jointly with feature distance to design anomaly scores. Instead of directly using the patch point clouds, we introduce adversarial perturbations to the input patch xyz data before feeding into the 3D normalcy classifier for the classification-based anomaly score. We show that 3DzAL outperforms the state-of-the-art anomaly detection and localization performance., Comment: This paper is accepted to WACV 2025

We present experimental results on tuning biphoton frequency by introducing a detuned coupling field in spontaneous four-wave mixing (SFWM), and examine its impact on the pairing ratio. This tunability is achieved by manipulating the inherent electromagnetically induced transparency (EIT) effect in the double-$\Lambda$ scheme. Introducing a detuned coupling field degrades the efficiency of EIT-based stimulated four-wave mixing, which in turn reduces the biphoton pairing ratio. However, this reduction can be mitigated by increasing the optical power of the coupling field. Additionally, we observe that blue- and red-detuning the biphoton frequency results in distinct temporal profiles of biphoton wavepackets due to phase mismatch. These findings provide insights into the mechanisms of frequency-tunable biphoton generation via SFWM, and suggest potential optimizations for applications in quantum communication and information processing.

This paper introduces a novel framework named D-LORD (Double Latent Optimization for Representation Disentanglement), which is designed for motion stylization (motion style transfer and motion retargeting). The primary objective of this framework is to separate the class and content information from a given motion sequence using a data-driven latent optimization approach. Here, class refers to person-specific style, such as a particular emotion or an individual's identity, while content relates to the style-agnostic aspect of an action, such as walking or jumping, as universally understood concepts. The key advantage of D-LORD is its ability to perform style transfer without needing paired motion data. Instead, it utilizes class and content labels during the latent optimization process. By disentangling the representation, the framework enables the transformation of one motion sequences style to another's style using Adaptive Instance Normalization. The proposed D-LORD framework is designed with a focus on generalization, allowing it to handle different class and content labels for various applications. Additionally, it can generate diverse motion sequences when specific class and content labels are provided. The framework's efficacy is demonstrated through experimentation on three datasets: the CMU XIA dataset for motion style transfer, the MHAD dataset, and the RRIS Ability dataset for motion retargeting. Notably, this paper presents the first generalized framework for motion style transfer and motion retargeting, showcasing its potential contributions in this area.

Galaxy-cluster virial (structure-formation accretion) shock observations are shown to imply $\gtrsim1\%$ magnetization of a layer extending $\gtrsim10^{16}$ Debye lengths downstream, challenging the modelling of high Alfv\'en-Mach collisionless shocks. Unlike similar shocks in supernova remnants or relativistic shocks in $\gamma$-ray burst afterglows, where macroscopic magnetized layers were detected but purportedly attributed to preexisting or non-resonant cosmic-ray streaming-seeded substructure, the upstream of strong virial shocks is both weakly magnetized and pristine. Hence, some mechanism must generate large-scale and possibly self-similar magnetic sub-structure out of the accreted primordial plasma; such a mechanism may dominate other high-Mach shock systems, too., Comment: 8 pages, 1 figure, comments welcome

Contact-rich bimanual manipulation involves precise coordination of two arms to change object states through strategically selected contacts and motions. Due to the inherent complexity of these tasks, acquiring sufficient demonstration data and training policies that generalize to unseen scenarios remain a largely unresolved challenge. Building on recent advances in planning through contacts, we introduce Generalizable Planning-Guided Diffusion Policy Learning (GLIDE), an approach that effectively learns to solve contact-rich bimanual manipulation tasks by leveraging model-based motion planners to generate demonstration data in high-fidelity physics simulation. Through efficient planning in randomized environments, our approach generates large-scale and high-quality synthetic motion trajectories for tasks involving diverse objects and transformations. We then train a task-conditioned diffusion policy via behavior cloning using these demonstrations. To tackle the sim-to-real gap, we propose a set of essential design options in feature extraction, task representation, action prediction, and data augmentation that enable learning robust prediction of smooth action sequences and generalization to unseen scenarios. Through experiments in both simulation and the real world, we demonstrate that our approach can enable a bimanual robotic system to effectively manipulate objects of diverse geometries, dimensions, and physical properties. Website: https://glide-manip.github.io/

In this study, the Quantum-Train Quantum Fast Weight Programmer (QT-QFWP) framework is proposed, which facilitates the efficient and scalable programming of variational quantum circuits (VQCs) by leveraging quantum-driven parameter updates for the classical slow programmer that controls the fast programmer VQC model. This approach offers a significant advantage over conventional hybrid quantum-classical models by optimizing both quantum and classical parameter management. The framework has been benchmarked across several time-series prediction tasks, including Damped Simple Harmonic Motion (SHM), NARMA5, and Simulated Gravitational Waves (GW), demonstrating its ability to reduce parameters by roughly 70-90\% compared to Quantum Long Short-term Memory (QLSTM) and Quantum Fast Weight Programmer (QFWP) without compromising accuracy. The results show that QT-QFWP outperforms related models in both efficiency and predictive accuracy, providing a pathway toward more practical and cost-effective quantum machine learning applications. This innovation is particularly promising for near-term quantum systems, where limited qubit resources and gate fidelities pose significant constraints on model complexity. QT-QFWP enhances the feasibility of deploying VQCs in time-sensitive applications and broadens the scope of quantum computing in machine learning domains., Comment: 9 pages, 7 figures

Surface electromyography (sEMG) recordings can be contaminated by electrocardiogram (ECG) signals when the monitored muscle is closed to the heart. Traditional signal-processing-based approaches, such as high-pass filtering and template subtraction, have been used to remove ECG interference but are often limited in their effectiveness. Recently, neural-network-based methods have shown greater promise for sEMG denoising, but they still struggle to balance both efficiency and effectiveness. In this study, we introduce MSEMG, a novel system that integrates the Mamba State Space Model with a convolutional neural network to serve as a lightweight sEMG denoising model. We evaluated MSEMG using sEMG data from the Non-Invasive Adaptive Prosthetics database and ECG signals from the MIT-BIH Normal Sinus Rhythm Database. The results show that MSEMG outperforms existing methods, generating higher-quality sEMG signals with fewer parameters. The source code for MSEMG is available at https://github.com/tonyliu0910/MSEMG., Comment: This paper is under review of 2025 ICASSP

The utilization of speech Self-Supervised Learning (SSL) models achieves impressive performance on Automatic Speech Recognition (ASR). However, in low-resource language ASR, they encounter the domain mismatch problem between pre-trained and low-resource languages. Typical solutions like fine-tuning the SSL model suffer from high computation costs while using frozen SSL models as feature extractors comes with poor performance. To handle these issues, we extend a conventional efficient fine-tuning scheme based on the adapter. We add an extra intermediate adaptation to warm up the adapter and downstream model initialization. Remarkably, we update only 1-5% of the total model parameters to achieve the adaptation. Experimental results on the ML-SUPERB dataset show that our solution outperforms conventional efficient fine-tuning. It achieves up to a 28% relative improvement in the Character/Phoneme error rate when adapting to unseen languages.

We investigate the tidal resonance of the fundamental ($f$-)mode in spinning neutron stars, robustly tracing the onset of the excitation to its saturation, using numerical relativity for the first time. We performed long-term ($\approx15$~orbits) fully relativistic simulations of a merger of two highly and retrogradely spinning neutron stars. The resonance window of the $f$-mode is extended by self-interaction, and the nonlinear resonance continues up to the final plunging phase. We observe that the quasi-circular orbit is maintained throughout since the dissipation of orbit motion due to the resonance is coherent with that due to gravitational waves. The $f$-mode resonance causes a variation in the stellar spin of $\gtrsim6.3\%$ in the linear regime and much more as $\sim33\%$ during the later nonlinear regime. At the merger, a phase shift of $\lesssim40$~radians is rendered in the gravitational waveform as a consequence of the angular momentum and energy transfers into the neutron star oscillations., Comment: 10 pages, 7 figures

Lane detection plays an important role in autonomous driving perception systems. As deep learning algorithms gain popularity, monocular lane detection methods based on them have demonstrated superior performance and emerged as a key research direction in autonomous driving perception. The core designs of these algorithmic frameworks can be summarized as follows: (1) Task paradigm, focusing on lane instance-level discrimination; (2) Lane modeling, representing lanes as a set of learnable parameters in the neural network; (3) Global context supplementation, enhancing inference on the obscure lanes; (4) Perspective effect elimination, providing accurate 3D lanes for downstream applications. From these perspectives, this paper presents a comprehensive overview of existing methods, encompassing both the increasingly mature 2D lane detection approaches and the developing 3D lane detection works. Besides, this paper compares the performance of mainstream methods on different benchmarks and investigates their inference speed under a unified setting for fair comparison. Moreover, we present some extended works on lane detection, including multi-task perception, video lane detection, online high-definition map construction, and lane topology reasoning, to offer readers a comprehensive roadmap for the evolution of lane detection. Finally, we point out some potential future research directions in this field. We exhaustively collect the papers and codes of existing works at https://github.com/Core9724/Awesome-Lane-Detection and will keep tracing the research.

Text-to-image (T2I) models have shown remarkable progress, but their potential to generate harmful content remains a critical concern in the ML community. While various safety mechanisms have been developed, the field lacks systematic tools for evaluating their effectiveness against real-world misuse scenarios. In this work, we propose ICER, a novel red-teaming framework that leverages Large Language Models (LLMs) and a bandit optimization-based algorithm to generate interpretable and semantic meaningful problematic prompts by learning from past successful red-teaming attempts. Our ICER efficiently probes safety mechanisms across different T2I models without requiring internal access or additional training, making it broadly applicable to deployed systems. Through extensive experiments, we demonstrate that ICER significantly outperforms existing prompt attack methods in identifying model vulnerabilities while maintaining high semantic similarity with intended content. By uncovering that successful jailbreaking instances can systematically facilitate the discovery of new vulnerabilities, our work provides crucial insights for developing more robust safety mechanisms in T2I systems.

This paper introduces the Global Challenge for Safe and Secure Large Language Models (LLMs), a pioneering initiative organized by AI Singapore (AISG) and the CyberSG R&D Programme Office (CRPO) to foster the development of advanced defense mechanisms against automated jailbreaking attacks. With the increasing integration of LLMs in critical sectors such as healthcare, finance, and public administration, ensuring these models are resilient to adversarial attacks is vital for preventing misuse and upholding ethical standards. This competition focused on two distinct tracks designed to evaluate and enhance the robustness of LLM security frameworks. Track 1 tasked participants with developing automated methods to probe LLM vulnerabilities by eliciting undesirable responses, effectively testing the limits of existing safety protocols within LLMs. Participants were challenged to devise techniques that could bypass content safeguards across a diverse array of scenarios, from offensive language to misinformation and illegal activities. Through this process, Track 1 aimed to deepen the understanding of LLM vulnerabilities and provide insights for creating more resilient models.

We provide and axiomatize a representation of preferences over lotteries that generalizes the expected utility model. Our representation is consistent with the violations of the independence axiom that we observe in the laboratory experiment that we conduct. The violations differ from the Allais Paradox in that they are incompatible with some of the most prominent non-expected utility models. Our representation can be interpreted as a decision-maker with context-dependent attitudes to risks and allows us to generate various types of realistic behavior. We analyze some properties of our model, including specifications that ensure preferences for first-order stochastic dominance. We test whether subjects in our experiment exhibit the type of context-dependent risk attitudes that arise in our model.

The integration of quantum computing into classical machine learning architectures has emerged as a promising approach to enhance model efficiency and computational capacity. In this work, we introduce the Quantum Kernel-Based Long Short-Term Memory (QK-LSTM) network, which utilizes quantum kernel functions within the classical LSTM framework to capture complex, non-linear patterns in sequential data. By embedding input data into a high-dimensional quantum feature space, the QK-LSTM model reduces the reliance on large parameter sets, achieving effective compression while maintaining accuracy in sequence modeling tasks. This quantum-enhanced architecture demonstrates efficient convergence, robust loss minimization, and model compactness, making it suitable for deployment in edge computing environments and resource-limited quantum devices (especially in the NISQ era). Benchmark comparisons reveal that QK-LSTM achieves performance on par with classical LSTM models, yet with fewer parameters, underscoring its potential to advance quantum machine learning applications in natural language processing and other domains requiring efficient temporal data processing.

Chest X-rays (CXRs) often display various diseases with disparate class frequencies, leading to a long-tailed, multi-label data distribution. In response to this challenge, we explore the Pruned MIMIC-CXR-LT dataset, a curated collection derived from the MIMIC-CXR dataset, specifically designed to represent a long-tailed and multi-label data scenario. We introduce LTCXNet, a novel framework that integrates the ConvNeXt model, ML-Decoder, and strategic data augmentation, further enhanced by an ensemble approach. We demonstrate that LTCXNet improves the performance of CXR interpretation across all classes, especially enhancing detection in rarer classes like `Pneumoperitoneum' and `Pneumomediastinum' by 79\% and 48\%, respectively. Beyond performance metrics, our research extends into evaluating fairness, highlighting that some methods, while improving model accuracy, could inadvertently affect fairness across different demographic groups negatively. This work contributes to advancing the understanding and management of long-tailed, multi-label data distributions in medical imaging, paving the way for more equitable and effective diagnostic tools., Comment: 8 pages, 5 figures

Motion capture technologies have transformed numerous fields, from the film and gaming industries to sports science and healthcare, by providing a tool to capture and analyze human movement in great detail. The holy grail in the topic of monocular global human mesh and motion reconstruction (GHMR) is to achieve accuracy on par with traditional multi-view capture on any monocular videos captured with a dynamic camera, in-the-wild. This is a challenging task as the monocular input has inherent depth ambiguity, and the moving camera adds additional complexity as the rendered human motion is now a product of both human and camera movement. Not accounting for this confusion, existing GHMR methods often output motions that are unrealistic, e.g. unaccounted root translation of the human causes foot sliding. We present DiffOpt, a novel 3D global HMR method using Diffusion Optimization. Our key insight is that recent advances in human motion generation, such as the motion diffusion model (MDM), contain a strong prior of coherent human motion. The core of our method is to optimize the initial motion reconstruction using the MDM prior. This step can lead to more globally coherent human motion. Our optimization jointly optimizes the motion prior loss and reprojection loss to correctly disentangle the human and camera motions. We validate DiffOpt with video sequences from the Electromagnetic Database of Global 3D Human Pose and Shape in the Wild (EMDB) and Egobody, and demonstrate superior global human motion recovery capability over other state-of-the-art global HMR methods most prominently in long video settings., Comment: 15 pages, 2 figures, submitted to TMLR

The Atacama Large Millimeter/submillimeter Array Survey of Orion Planck Galactic Cold Clumps (ALMASOP) reveals complex nested morphological and kinematic features of molecular outflows through the CO (J = 2 - 1) and SiO (J = 5 - 4) emission. We characterize the jet and outflow kinematics of the ALMASOP sample in four representative sources (HOPS 10, 315, 358, and G203.21-11.20W2) through channel maps and position-velocity diagrams (PVDs) parallel and transverse to the outflow axes. The combined CO and SiO emission exhibits the coexistence of the conventional extremely-high-velocity (EHV) jets and shell-like low-velocity (LV) cavity walls and new features. More complex, nested bubble-like and filamentary structures in the images and channel maps, triangle-shaped regions near the base of the parallel PVDs, and regions composed of rhombus/oval shapes in the transverse PVDs, are also evident. Such features find natural explanations within the bubble structure of the unified model of jet, wind, and ambient medium. The reverse shock cavity is revealed on the PVD base regions, and other features naturally arise within the dynamic postshock region of magnetic interaction. The finer nested shells observed within the compressed wind region reveal previously unnoticed shocked emission between the jet and the conventional large cavity walls. These pseudopulse-produced filamentary features connect to the jet-like knotty blobs, creating an impression of episodicity in mass ejection. SiO emission is enhanced downstream of the reverse shock boundary, with jet-like excitation conditions. Combined, these observed features reveal the extended structures induced by the magnetic interplay between a jet-bearing magnetized wide-angle wind and its ambient magnetized surrounding medium., Comment: 27 pages, 11 figures, accepted by ApJ

Continual learning (CL) is crucial for language models to dynamically adapt to the evolving real-world demands. To mitigate the catastrophic forgetting problem in CL, data replay has been proven a simple and effective strategy, and the subsequent data-replay-based distillation can further enhance the performance. However, existing methods fail to fully exploit the knowledge embedded in models from previous tasks, resulting in the need for a relatively large number of replay samples to achieve good results. In this work, we first explore and emphasize the importance of attention weights in knowledge retention, and then propose a SElective attEntion-guided Knowledge Retention method (SEEKR) for data-efficient replay-based continual learning of large language models (LLMs). Specifically, SEEKR performs attention distillation on the selected attention heads for finer-grained knowledge retention, where the proposed forgettability-based and task-sensitivity-based measures are used to identify the most valuable attention heads. Experimental results on two continual learning benchmarks for LLMs demonstrate the superiority of SEEKR over the existing methods on both performance and efficiency. Explicitly, SEEKR achieves comparable or even better performance with only 1/10 of the replayed data used by other methods, and reduces the proportion of replayed data to 1%., Comment: EMNLP2024

Oblivious RAM (ORAM) hides the memory access patterns, enhancing data privacy by preventing attackers from discovering sensitive information based on the sequence of memory accesses. The performance of ORAM is often limited by its inherent trade-off between security and efficiency, as concealing memory access patterns imposes significant computational and memory overhead. While prior works focus on improving the ORAM performance by prefetching and eliminating ORAM requests, we find that their performance is very sensitive to workload locality behavior and incurs additional management overhead caused by the ORAM stash pressure. This paper presents Palermo: a protocol-hardware co-design to improve ORAM performance. The key observation in Palermo is that classical ORAM protocols enforce restrictive dependencies between memory operations that result in low memory bandwidth utilization. Palermo introduces a new protocol that overlaps large portions of memory operations, within a single and between multiple ORAM requests, without breaking correctness and security guarantees. Subsequently, we propose an ORAM controller architecture that executes the proposed protocol to service ORAM requests. The hardware is responsible for concurrently issuing memory requests as well as imposing the necessary dependencies to ensure a consistent view of the ORAM tree across requests. Using a rich workload mix, we demonstrate that Palermo outperforms the RingORAM baseline by 2.8x, on average, incurring a negligible area overhead of 5.78mm^2 (less than 2% in 12th generation Intel CPU after technology scaling) and 2.14W without sacrificing security. We further show that Palermo also outperforms the state-of-the-art works PageORAM, PrORAM, and IR-ORAM., Comment: To appear in HPCA'25

Quantum magnetism is one of the most active fields for exploring exotic phases and phase transitions. The recently synthesized Na2BaCo(PO4)2 (NBCP) is an ideal material incarnation of the spin-1/2 easy axis triangular lattice antiferromagnet (TLAF). Experimental evidence shows that NBCP hosts the spin supersolid state with a giant magnetocaloric effect. It was also proposed that the applied magnetic field B can drive the system through Berezinskii-Kosterlitz-Thouless (BKT) and other richer quantum phase transitions. However, the detection of these transitions is challenging because they onset at extremely low temperature T at around 60 mK, and the measurement of the magnetic susceptibility of these transitions requires high sensitivity. With the help of our newly developed gradient force magnetometer in a dilution refrigerator, we constructed the contour diagram of the magnetic susceptibility in the B-T phase diagram in T as cold as 30 mK. These results provide a more comprehensive and accurate understanding of the several field-tunable quantum phase transitions and BKT melting of the spin supersolidity, which are especially significant when their giant magnetocaloric effects highlight potential applications for sub-Kelvin refrigeration under concerns about global helium shortages., Comment: 19 pages, 5 main figures, and 4 figures in the supplement

We present efficient counting and sampling algorithms for random $k$-SAT when the clause density satisfies $\alpha \le \frac{2^k}{\mathrm{poly}(k)}.$ In particular, the exponential term $2^k$ matches the satisfiability threshold $\Theta(2^k)$ for the existence of a solution and the (conjectured) algorithmic threshold $2^k (\ln k) / k$ for efficiently finding a solution. Previously, the best-known counting and sampling algorithms required far more restricted densities $\alpha\lesssim 2^{k/3}$ [He, Wu, Yang, SODA '23]. Notably, our result goes beyond the lower bound $d\gtrsim 2^{k/2}$ for worst-case $k$-SAT with bounded-degree $d$ [Bez\'akov\'a et al, SICOMP '19], showing that for counting and sampling, the average-case random $k$-SAT model is computationally much easier than the worst-case model. At the heart of our approach is a new refined analysis of the recent novel coupling procedure by [Wang, Yin, FOCS '24], utilizing the structural properties of random constraint satisfaction problems (CSPs). Crucially, our analysis avoids reliance on the $2$-tree structure used in prior works, which cannot extend beyond the worst-case threshold $2^{k/2}$. Instead, we employ a witness tree similar to that used in the analysis of the Moser-Tardos algorithm [Moser, Tardos, JACM '10] for the Lov\'{a}sz Local lemma, which may be of independent interest. Our new analysis provides a universal framework for efficient counting and sampling for random atomic CSPs, including, for example, random hypergraph colorings. At the same time, it immediately implies as corollaries several structural and probabilistic properties of random CSPs that have been widely studied but rarely justified, including replica symmetry and non-reconstruction., Comment: 39 pages

This paper considers a patrol inspection scenario where multiple unmanned aerial vehicles (UAVs) are adopted to traverse multiple predetermined cruise points for data collection. The UAVs are connected to cellular networks and they would offload the collected data to the ground base stations (GBSs) for data processing within the constrained duration. This paper proposes a balanced task assignment strategy among patrol UAVs and an energy-efficient trajectory design method. Through jointly optimizing the cruise point assignment, communication scheduling, computational allocation, and UAV trajectory, a novel solution can be obtained to balance the multiple UAVs' task completion time and minimize the total energy consumption. Firstly, we propose a novel clustering method that considers geometry topology, communication rate, and offload volume; it can determine each UAV's cruise points and balance the UAVs' patrol task. Secondly, a hybrid Time-Energy traveling salesman problem is formulated to analyze the cruise point traversal sequence, and the energy-efficient UAV trajectory can be designed by adopting the successive convex approximation (SCA) technique and block coordinate descent (BCD) scheme. The numerical results demonstrate that the proposed balanced task assignment strategy can efficiently balance the multiple UAVs' tasks. Moreover, the min-max task completion time and total energy consumption performance of the proposed solution outperform that of the current conventional approach.

The trade-off effect of the oxygen permeability and stability of oxygen transport membranes (OTMs) still exists in working atmospheres containing CO2. Herein, we reported a new series of 60 wt%Ce0.9Ln0.1O2-{\delta}-40wt%Ln0.6Sr0.4Fe0.9Ti0.1O3-{\delta} (CLnO-LnSFTO, Ln = La, Pr, Nd, Sm, Gd, Tb) dual-phase OTMs by selecting different Ln elements based on the reported highly stable Ti-doped CPrO-PrSFTO. The effects of different Ln elements on the structure and oxygen permeability of Ti-doped dual-phase OTMs were systematically studied. Basically, as the atomic number of Ln elements increases, the unit cell parameters of both the fluorite phase and the perovskite phase become smaller. The unit cell volume and spatial symmetry of the perovskite phase are reduced, resulting in a reduction in oxygen permeability. The optimal CLaO-LaSFTO showed JO2 of 0.60 and 0.54 mL min-1 cm-2 with He and CO2 sweeping at 1000 oC, respectively. In addition, all CLnO-LnSFTO OTMs could work for more than 100 hours with no significant performance degradation in a CO2 atmosphere, maintaining excellent stability. This work explores candidate OTM materials for CO2 capture and oxygen separation, as well as provides some ideas for addressing the trade-off effect., Comment: 28 pages, 7 figures

We present a novel approach for reconstructing annual temperatures in East Asia from 1368 to 1911, leveraging the Reconstructed East Asian Climate Historical Encoded Series (REACHES). The lack of instrumental data during this period poses significant challenges to understanding past climate conditions. REACHES digitizes historical documents from the Ming and Qing dynasties of China, converting qualitative descriptions into a four-level ordinal temperature scale. However, these index-based data are biased toward abnormal or extreme weather phenomena, leading to data gaps that likely correspond to normal conditions. To address this bias and reconstruct historical temperatures at any point within East Asia, including locations without direct historical data, we employ a three-tiered statistical framework. First, we perform kriging to interpolate temperature data across East Asia, adopting a zero-mean assumption to handle missing information. Next, we utilize the Last Millennium Ensemble (LME) reanalysis data and apply quantile mapping to calibrate the kriged REACHES data to Celsius temperature scales. Finally, we introduce a novel Bayesian data assimilation method that integrates the kriged Celsius data with LME simulations to enhance reconstruction accuracy. We model the LME data at each geographic location using a flexible nonstationary autoregressive time series model and employ regularized maximum likelihood estimation with a fused lasso penalty. The resulting dynamic distribution serves as a prior, which is refined via Kalman filtering by incorporating the kriged Celsius REACHES data to yield posterior temperature estimates. This comprehensive integration of historical documentation, contemporary climate models, and advanced statistical methods improves the accuracy of historical temperature reconstructions and provides a crucial resource for future environmental and climate studies., Comment: 31 pages, 15 figures, 1 table

Most learned B-frame codecs with hierarchical temporal prediction suffer from the domain shift issue caused by the discrepancy in the Group-of-Pictures (GOP) size used for training and test. As such, the motion estimation network may fail to predict large motion properly. One effective strategy to mitigate this domain shift issue is to downsample video frames for motion estimation. However, finding the optimal downsampling factor involves a time-consuming rate-distortion optimization process. This work introduces lightweight classifiers to determine the downsampling factor. To strike a good rate-distortion-complexity trade-off, our classifiers observe simple state signals, including only the coding and reference frames, to predict the best downsampling factor. We present two variants that adopt binary and multi-class classifiers, respectively. The binary classifier adopts the Focal Loss for training, classifying between motion estimation at high and low resolutions. Our multi-class classifier is trained with novel soft labels incorporating the knowledge of the rate-distortion costs of different downsampling factors. Both variants operate as add-on modules without the need to re-train the B-frame codec. Experimental results confirm that they achieve comparable coding performance to the brute-force search methods while greatly reducing computational complexity.

The spatial distribution of spurious dots in SiMOS single-electron transistors (SETs), fabricated on an industrial 300 mm process line, has been statistically analyzed. To have a deeper understanding of the origin of these spurious dots, we analyzed SETs with three different oxide thicknesses: 8 nm, 12 nm and 20 nm. By combining spurious dot triangulation cryo-measurement with simulations of strain, gate bias, and location of the electron wave function, we demonstrate that most spurious dots are formed through the combined effects of strain and gate bias, leading to variations in the conduction band energy. Despite the similar thermal expansion coefficients of polycrystalline silicon gates and single-crystalline silicon substrates, strain remains a crucial factor in spurious dots formation. This learning can be use to optimize the device design and the oxide thickness, to reduce the density of spurious dot while keeping quantum dot tunability., Comment: 10 pages, 9 figures

Single-shot ptychography is a quantitative phase imaging method wherein overlapping beams of light arranged in a grid pattern simultaneously illuminate a sample, allowing a full ptychographic dataset to be collected in a single shot. It is primarily used at optical wavelengths, but there is interest in using it for X-ray imaging. However, the constraints imposed by X-ray optics have limited the resolution achievable to date. In this work, we reinterpret single-shot ptychography as a structured illumination method by viewing the grid of beams as a single, highly structured illumination function. Pre-calibrating this illumination and reconstructing single-shot data using the randomized probe imaging algorithm allows us to account for the overlap and coherent interference between the diffraction arising from each beam. We achieve a resolution 3.5 times finer than the numerical aperture-based limit imposed by traditional algorithms for single-shot ptychography. We argue that this reconstruction method will work better for most single-shot ptychography experiments and discuss the implications for the design of future single-shot X-ray microscopes., Comment: 4 pages, 3 figures

Recent advancements in large audio-language models (LALMs) have shown impressive capabilities in understanding and reasoning about audio and speech information. However, these models still face challenges, including hallucinating non-existent sound events, misidentifying the order of sound events, and incorrectly attributing sound sources, which undermine their reliability and real-world application. To systematically evaluate these issues, we propose three distinct tasks: object existence, temporal order, and object attribute within audio. These tasks assess the models' comprehension of critical audio information aspects. Our experimental results reveal limitations in these fundamental tasks, underscoring the need for better models in recognizing specific sound events, determining event sequences, and identifying sound sources. To improve performance in these areas, we introduce a multi-turn chain-of-thought approach, which demonstrates significantly improved model performance across the proposed tasks., Comment: 5 pages, 1 figure

Interactions of isolated quantum many-body systems typically scramble local information into the entire system and make it unrecoverable. Ergodicity-breaking systems possess the potential to exhibit fundamentally different information scrambling dynamics beyond this paradigm. For many-body localized systems with strong ergodicity breaking, local transport vanishes and information scrambles logarithmically slowly. Whereas in Rydberg atom arrays, local qubit flips induce dynamical retardation on surrounding qubits through the Rydberg blockade effect, giving rise to quantum many-body scars that weakly break ergodicity, and resulting in the predicted unconventional quantum information spreading behaviours. Here, we present the first measurements of out-of-time-ordered correlators and Holevo information in a Rydberg atom array, enabling us to precisely track quantum information scrambling and transport dynamics. By leveraging these tools, we observe a novel spatio-temporal collapse-and-revival behaviour of quantum information, which differs from both typical chaotic and many-body localized systems. Our experiment sheds light on the unique information dynamics in many-body systems with kinetic constraints, and demonstrates an effective digital-analogue approach to coherently reverse time evolution and steer information propagation in near-term quantum devices., Comment: 12 pages, 6 figures + Supplementary Information 37 pages, 24 figures

Chemical doping is a critical factor in the development of new superconductors or optimizing the superconducting transition temperature (Tc) of the parent superconducting materials. Herein, a new simple urea approach is developed to synthesize the N-doped alfa-Mo2C. Benefiting from the simple urea method, a broad superconducting dome is found in the Mo2C1-xNx compositions. XRD results show that the structure of alfa-Mo2C remains unchanged and that there is a variation of lattice parameters with nitrogen doping. Resistivity, magnetic susceptibility, and heat capacity measurement results confirm that the superconducting transition temperature (Tc) was strongly increased from 2.68 K (x = 0) to 7.05 K (x = 0.49). First-principles calculations and our analysis indicate that increasing nitrogen doping leads to a rise in the density of states at the Fermi level and doping-induced phonon softening, which enhances electron-phonon coupling. This results in an increase in Tc and a sharp rise in the upper critical field. Our findings provide a promising strategy for fabricating transition metal carbonitrides and provide a material platform for further study of the superconductivity of transition metal carbides., Comment: 15 pages, 6 Figures, 1 Table

Off-axis digital holographic microscopy is a high-throughput, label-free imaging technology that provides three-dimensional, high-resolution information about samples, particularly useful in large-scale cellular imaging. However, the hologram reconstruction process poses a significant bottleneck for timely data analysis. To address this challenge, we propose a novel reconstruction approach that integrates deep learning with the physical principles of off-axis holography. We initialized part of the network weights based on the physical principle and then fine-tuned them via weakly supersized learning. Our off-axis hologram network (OAH-Net) retrieves phase and amplitude images with errors that fall within the measurement error range attributable to hardware, and its reconstruction speed significantly surpasses the microscope's acquisition rate. Crucially, OAH-Net demonstrates remarkable external generalization capabilities on unseen samples with distinct patterns and can be seamlessly integrated with other models for downstream tasks to achieve end-to-end real-time hologram analysis. This capability further expands off-axis holography's applications in both biological and medical studies., Comment: 11 pages, 4 figures

Multi-camera systems are indispensable in movies, TV shows, and other media. Selecting the appropriate camera at every timestamp has a decisive impact on production quality and audience preferences. Learning-based view recommendation frameworks can assist professionals in decision-making. However, they often struggle outside of their training domains. The scarcity of labeled multi-camera view recommendation datasets exacerbates the issue. Based on the insight that many videos are edited from the original multi-camera videos, we propose transforming regular videos into pseudo-labeled multi-camera view recommendation datasets. Promisingly, by training the model on pseudo-labeled datasets stemming from videos in the target domain, we achieve a 68% relative improvement in the model's accuracy in the target domain and bridge the accuracy gap between in-domain and never-before-seen domains., Comment: Accepted to VCIP 2024. Project page: https://eric11220.github.io/publication/VCIP24/

Domain-specific adaptation is critical to maximizing the performance of pre-trained language models (PLMs) on one or multiple targeted tasks, especially under resource-constrained use cases, such as edge devices. However, existing methods often struggle to balance domain-specific performance, retention of general knowledge, and efficiency for training and inference. To address these challenges, we propose Modular Domain Experts (MoDE). MoDE is a mixture-of-experts architecture that augments a general PLMs with modular, domain-specialized experts. These experts are trained independently and composed together via a lightweight training process. In contrast to standard low-rank adaptation methods, each MoDE expert consists of several transformer layers which scale better with more training examples and larger parameter counts. Our evaluation demonstrates that MoDE achieves comparable target performances to full parameter fine-tuning while achieving 1.65% better retention performance. Moreover, MoDE's architecture enables flexible sharding configurations and improves training speeds by up to 38% over state-of-the-art distributed training configurations., Comment: 14 pages, 5 figures, 3 tables

The rise of deepfake technologies has posed significant challenges to privacy, security, and information integrity, particularly in audio and multimedia content. This paper introduces a Quantum-Trained Convolutional Neural Network (QT-CNN) framework designed to enhance the detection of deepfake audio, leveraging the computational power of quantum machine learning (QML). The QT-CNN employs a hybrid quantum-classical approach, integrating Quantum Neural Networks (QNNs) with classical neural architectures to optimize training efficiency while reducing the number of trainable parameters. Our method incorporates a novel quantum-to-classical parameter mapping that effectively utilizes quantum states to enhance the expressive power of the model, achieving up to 70% parameter reduction compared to classical models without compromising accuracy. Data pre-processing involved extracting essential audio features, label encoding, feature scaling, and constructing sequential datasets for robust model evaluation. Experimental results demonstrate that the QT-CNN achieves comparable performance to traditional CNNs, maintaining high accuracy during training and testing phases across varying configurations of QNN blocks. The QT framework's ability to reduce computational overhead while maintaining performance underscores its potential for real-world applications in deepfake detection and other resource-constrained scenarios. This work highlights the practical benefits of integrating quantum computing into artificial intelligence, offering a scalable and efficient approach to advancing deepfake detection technologies.

The piecewise-stationary bandit problem is an important variant of the multi-armed bandit problem that further considers abrupt changes in the reward distributions. The main theme of the problem is the trade-off between exploration for detecting environment changes and exploitation of traditional bandit algorithms. While this problem has been extensively investigated, existing works either assume knowledge about the number of change points $M$ or require extremely high computational complexity. In this work, we revisit the piecewise-stationary bandit problem from a minimalist perspective. We propose a novel and generic exploration mechanism, called diminishing exploration, which eliminates the need for knowledge about $M$ and can be used in conjunction with an existing change detection-based algorithm to achieve near-optimal regret scaling. Simulation results show that despite oblivious of $M$, equipping existing algorithms with the proposed diminishing exploration generally achieves better empirical regret than the traditional uniform exploration.

Polarization measurements provide insight into the magnetic field, a critical aspect of the dynamics and emission properties around the compact object. In this paper, we present the polarized magnetic field of the Crab outer torus and the Vela arc utilizing Imaging X-ray Polarimetry Explorer observation data. The polarization angle (PA) measured for the Crab outer torus exhibits two monotonic evolutions along the azimuth angle, which are consistent with the normal line of the elliptical ring. There is a slight increase in PA along the azimuth angle for both the inner arc and the outer arc of the Vela nebula. The polarized magnetic vector along the outer torus of the Crab nebula shows the polarized magnetic field aligns with Crab outer torus structure. The PA variation along the Crab outer torus suggests a bulk flow speed of 0.8c. Meanwhile, the Vela nebula polarized magnetic field does not exactly align with the Vela arc structure. We noted that the Crab nebula possesses a polarized toroidal magnetic field, where as the Vela nebula exhibits an incomplete toroidal magnetic field., Comment: 8 pages, 5 figures

The recent development of Video-based Large Language Models (VideoLLMs), has significantly advanced video summarization by aligning video features and, in some cases, audio features with Large Language Models (LLMs). Each of these VideoLLMs possesses unique strengths and weaknesses. Many recent methods have required extensive fine-tuning to overcome the limitations of these models, which can be resource-intensive. In this work, we observe that the strengths of one VideoLLM can complement the weaknesses of another. Leveraging this insight, we propose a novel video summarization framework inspired by the Mixture of Experts (MoE) paradigm, which operates as an inference-time algorithm without requiring any form of fine-tuning. Our approach integrates multiple VideoLLMs to generate comprehensive and coherent textual summaries. It effectively combines visual and audio content, provides detailed background descriptions, and excels at identifying keyframes, which enables more semantically meaningful retrieval compared to traditional computer vision approaches that rely solely on visual information, all without the need for additional fine-tuning. Moreover, the resulting summaries enhance performance in downstream tasks such as summary video generation, either through keyframe selection or in combination with text-to-image models. Our language-driven approach offers a semantically rich alternative to conventional methods and provides flexibility to incorporate newer VideoLLMs, enhancing adaptability and performance in video summarization tasks.

In vision-language models (VLMs), visual tokens usually consume a significant amount of computational overhead, despite their sparser information density compared to text tokens. To address this, most existing methods learn a network to prune redundant visual tokens and require additional training data. Differently, we propose an efficient training-free token optimization mechanism dubbed SparseVLM without extra parameters or fine-tuning costs. Concretely, given that visual tokens complement text tokens in VLMs for linguistic reasoning, we select visual-relevant text tokens to rate the significance of vision tokens within the self-attention matrix extracted from the VLMs. Then we progressively prune irrelevant tokens. To maximize sparsity while retaining essential information, we introduce a rank-based strategy to adaptively determine the sparsification ratio for each layer, alongside a token recycling method that compresses pruned tokens into more compact representations. Experimental results show that our SparseVLM improves the efficiency of various VLMs across a range of image and video understanding tasks. In particular, LLaVA equipped with SparseVLM reduces 61% to 67% FLOPs with a compression ratio of 78% while maintaining 93% of the accuracy. Our code is available at https://github.com/Gumpest/SparseVLMs., Comment: 17 pages

Surface electromyography (sEMG) is a widely employed bio-signal that captures human muscle activity via electrodes placed on the skin. Several studies have proposed methods to remove sEMG contaminants, as non-invasive measurements render sEMG susceptible to various contaminants. However, these approaches often rely on heuristic-based optimization and are sensitive to the contaminant type. A more potent, robust, and generalized sEMG denoising approach should be developed for various healthcare and human-computer interaction applications. This paper proposes a novel neural network (NN)-based sEMG denoising method called TrustEMG-Net. It leverages the potent nonlinear mapping capability and data-driven nature of NNs. TrustEMG-Net adopts a denoising autoencoder structure by combining U-Net with a Transformer encoder using a representation-masking approach. The proposed approach is evaluated using the Ninapro sEMG database with five common contamination types and signal-to-noise ratio (SNR) conditions. Compared with existing sEMG denoising methods, TrustEMG-Net achieves exceptional performance across the five evaluation metrics, exhibiting a minimum improvement of 20%. Its superiority is consistent under various conditions, including SNRs ranging from -14 to 2 dB and five contaminant types. An ablation study further proves that the design of TrustEMG-Net contributes to its optimality, providing high-quality sEMG and serving as an effective, robust, and generalized denoising solution for sEMG applications., Comment: 18 pages, 7 figures, to be published in IEEE Journal of Biomedical and Health Informatics

This paper aims to delve into the rate-distortion-complexity trade-offs of modern neural video coding. Recent years have witnessed much research effort being focused on exploring the full potential of neural video coding. Conditional autoencoders have emerged as the mainstream approach to efficient neural video coding. The central theme of conditional autoencoders is to leverage both spatial and temporal information for better conditional coding. However, a recent study indicates that conditional coding may suffer from information bottlenecks, potentially performing worse than traditional residual coding. To address this issue, recent conditional coding methods incorporate a large number of high-resolution features as the condition signal, leading to a considerable increase in the number of multiply-accumulate operations, memory footprint, and model size. Taking DCVC as the common code base, we investigate how the newly proposed conditional residual coding, an emerging new school of thought, and its variants may strike a better balance among rate, distortion, and complexity., Comment: Accepted to MMSP 2024

Chronic wounds and associated complications present ever growing burdens for clinics and hospitals world wide. Venous, arterial, diabetic, and pressure wounds are becoming increasingly common globally. These conditions can result in highly debilitating repercussions for those affected, with limb amputations and increased mortality risk resulting from infection becoming more common. New methods to assist clinicians in chronic wound care are therefore vital to maintain high quality care standards. This paper presents an improved HarDNet segmentation architecture which integrates a contrast-eliminating component in the initial layers of the network to enhance feature learning. We also utilise a multi-colour space tensor merging process and adjust the harmonic shape of the convolution blocks to facilitate these additional features. We train our proposed model using wound images from light-skinned patients and test the model on two test sets (one set with ground truth, and one without) comprising only darker-skinned cases. Subjective ratings are obtained from clinical wound experts with intraclass correlation coefficient used to determine inter-rater reliability. For the dark-skin tone test set with ground truth, we demonstrate improvements in terms of Dice similarity coefficient (+0.1221) and intersection over union (+0.1274). Qualitative analysis showed high expert ratings, with improvements of >3% demonstrated when comparing the baseline model with the proposed model. This paper presents the first study to focus on darker-skin tones for chronic wound segmentation using models trained only on wound images exhibiting lighter skin. Diabetes is highly prevalent in countries where patients have darker skin tones, highlighting the need for a greater focus on such cases. Additionally, we conduct the largest qualitative study to date for chronic wound segmentation.

While fine-tuning pretrained models has become common practice, these models often underperform outside their specific domains. Recently developed model merging techniques enable the direct integration of multiple models, each fine-tuned for distinct tasks, into a single model. This strategy promotes multitasking capabilities without requiring retraining on the original datasets. However, existing methods fall short in addressing potential conflicts and complex correlations between tasks, especially in parameter-level adjustments, posing a challenge in effectively balancing parameter competition across various tasks. This paper introduces an innovative technique named PCB-Merging (Parameter Competition Balancing), a lightweight and training-free technique that adjusts the coefficients of each parameter for effective model merging. PCB-Merging employs intra-balancing to gauge parameter significance within individual tasks and inter-balancing to assess parameter similarities across different tasks. Parameters with low importance scores are dropped, and the remaining ones are rescaled to form the final merged model. We assessed our approach in diverse merging scenarios, including cross-task, cross-domain, and cross-training configurations, as well as out-of-domain generalization. The experimental results reveal that our approach achieves substantial performance enhancements across multiple modalities, domains, model sizes, number of tasks, fine-tuning forms, and large language models, outperforming existing model merging methods. The code is publicly available at: \url{https://github.com/duguodong7/pcb-merging}., Comment: Accepted by NeurIPS2024

We introduce Functional Group-Aware Representations for Small Molecules (FARM), a novel foundation model designed to bridge the gap between SMILES, natural language, and molecular graphs. The key innovation of FARM lies in its functional group-aware tokenization, which directly incorporates functional group information into the representations. This strategic reduction in tokenization granularity is intentionally aligned with key drivers of functional properties (i.e., functional groups), enhancing the model's understanding of chemical language. By expanding the chemical lexicon, FARM more effectively bridges SMILES and natural language, ultimately advancing the model's capacity to predict molecular properties. FARM also represents molecules from two perspectives: by using masked language modeling to capture atom-level features and by employing graph neural networks to encode the whole molecule topology. By leveraging contrastive learning, FARM aligns these two views of representations into a unified molecular embedding. We rigorously evaluate FARM on the MoleculeNet dataset, where it achieves state-of-the-art performance on 10 out of 12 tasks. These results highlight FARM's potential to improve molecular representation learning, with promising applications in drug discovery and pharmaceutical research., Comment: Preprint

The Ly$\alpha$ emission has emerged as a powerful tool for probing diffuse gas within the large-scale structure of the universe. In this paper, we investigate cosmic Ly$\alpha$ emission by post-processing cosmological simulations from \texttt{IllustrisTNG} and \texttt{THESAN} project. Specifically, we calculate the Ly$\alpha$ emission from galaxies, circum-galactic medium (CGM) and inter-galactic medium (IGM) across various redshifts. Our results show that IGM alone is significantly under the current observational upper limits. Meanwhile, CGM overshoots the observed galaxy contribution at $z \lesssim 0.5$ indicating that either the escape fraction for the inner CGM is less than unity or the current photoionization equilibrium treatment with an approximate self-shielding prescription is less accurate. The galaxy component also overshoots at low redshift, indicating that the escape fraction has strong evolution caused by an evolving halo mass function and dust growth distribution, that agrees with observationally inferred escape fractions. Furthermore, our findings suggest that the Ly$\alpha$ emission from diffuse gas (CGM+IGM) peaked at $z \sim 4$ and diminishes toward lower redshift. The Ly$\alpha$ emission from diffuse gas mainly originates through the collisional excitation of hot plasma. By comparing models with observation, our predicted Ly$\alpha$ emission from diffuse gas remains $\sim 6$ times fainter than the observed cosmic Ly$\alpha$ emission at $z=1-3$. However, future large telescopes may hold great promise to detect Ly$\alpha$ emission from diffuse gas toward $z>3$.