Search

3D Gaussian Splatting is emerging as a state-of-the-art technique in novel view synthesis, recognized for its impressive balance between visual quality, speed, and rendering efficiency. However, reliance on third-degree spherical harmonics for color representation introduces significant storage demands and computational overhead, resulting in a large memory footprint and slower rendering speed. We introduce SG-Splatting with Spherical Gaussians based color representation, a novel approach to enhance rendering speed and quality in novel view synthesis. Our method first represents view-dependent color using Spherical Gaussians, instead of three degree spherical harmonics, which largely reduces the number of parameters used for color representation, and significantly accelerates the rendering process. We then develop an efficient strategy for organizing multiple Spherical Gaussians, optimizing their arrangement to achieve a balanced and accurate scene representation. To further improve rendering quality, we propose a mixed representation that combines Spherical Gaussians with low-degree spherical harmonics, capturing both high- and low-frequency color information effectively. SG-Splatting also has plug-and-play capability, allowing it to be easily integrated into existing systems. This approach improves computational efficiency and overall visual fidelity, making it a practical solution for real-time applications.

We introduce SeaDAG, a semi-autoregressive diffusion model for conditional generation of Directed Acyclic Graphs (DAGs). Considering their inherent layer-wise structure, we simulate layer-wise autoregressive generation by designing different denoising speed for different layers. Unlike conventional autoregressive generation that lacks a global graph structure view, our method maintains a complete graph structure at each diffusion step, enabling operations such as property control that require the full graph structure. Leveraging this capability, we evaluate the DAG properties during training by employing a graph property decoder. We explicitly train the model to learn graph conditioning with a condition loss, which enhances the diffusion model's capacity to generate graphs that are both realistic and aligned with specified properties. We evaluate our method on two representative conditional DAG generation tasks: (1) circuit generation from truth tables, where precise DAG structures are crucial for realizing circuit functionality, and (2) molecule generation based on quantum properties. Our approach demonstrates promising results, generating high-quality and realistic DAGs that closely align with given conditions.

Fabrication of dielectrics at low temperature is required for temperature-sensitive detectors. For superconducting detectors, such as transition edge sensors and kinetic inductance detectors, AlMn is widely studied due to its variable superconducting transition temperature at different baking temperatures. Experimentally only the highest baking temperature determines AlMn transition temperature, so we need to control the wafer temperature during the whole process. In general, the highest process temperature happens during dielectric fabrication. Here, we present the cryogenic microwave performance of Si$_{3}$N$_{4}$, SiN$_{x}$ and $\alpha$-Si using ICPCVD at low temperature of 75 $^{\circ}$C. The dielectric constant, internal quality factor and TLS properties are studied using Al parallel plate resonators.

Target speech extraction (TSE) focuses on extracting the speech of a specific target speaker from a mixture of signals. Existing TSE models typically utilize static embeddings as conditions for extracting the target speaker's voice. However, the static embeddings often fail to capture the contextual information of the extracted speech signal, which may limit the model's performance. We propose a novel dynamic embedding causal target speech extraction model to address this limitation. Our approach incorporates an autoregressive mechanism to generate context-dependent embeddings based on the extracted speech, enabling real-time, frame-level extraction. Experimental results demonstrate that the proposed model enhances short-time objective intelligibility (STOI) and signal-to-distortion ratio (SDR), offering a promising solution for target speech extraction in challenging scenarios., Comment: The experimental design and results contain errors, and I would like to withdraw the paper

The Transformer model, particularly its cross-attention module, is widely used for feature fusion in target sound extraction which extracts the signal of interest based on given clues. Despite its effectiveness, this approach suffers from low computational efficiency. Recent advancements in state space models, notably the latest work Mamba, have shown comparable performance to Transformer-based methods while significantly reducing computational complexity in various tasks. However, Mamba's applicability in target sound extraction is limited due to its inability to capture dependencies between different sequences as the cross-attention does. In this paper, we propose CrossMamba for target sound extraction, which leverages the hidden attention mechanism of Mamba to compute dependencies between the given clues and the audio mixture. The calculation of Mamba can be divided to the query, key and value. We utilize the clue to generate the query and the audio mixture to derive the key and value, adhering to the principle of the cross-attention mechanism in Transformers. Experimental results from two representative target sound extraction methods validate the efficacy of the proposed CrossMamba., Comment: 5 pages, 2 figures, accepted by ICASSP 2025

Register Transfer Level (RTL) code optimization is crucial for enhancing the efficiency and performance of digital circuits during early synthesis stages. Currently, optimization relies heavily on manual efforts by skilled engineers, often requiring multiple iterations based on synthesis feedback. In contrast, existing compiler-based methods fall short in addressing complex designs. This paper introduces RTLRewriter, an innovative framework that leverages large models to optimize RTL code. A circuit partition pipeline is utilized for fast synthesis and efficient rewriting. A multi-modal program analysis is proposed to incorporate vital visual diagram information as optimization cues. A specialized search engine is designed to identify useful optimization guides, algorithms, and code snippets that enhance the model ability to generate optimized RTL. Additionally, we introduce a Cost-aware Monte Carlo Tree Search (C-MCTS) algorithm for efficient rewriting, managing diverse retrieved contents and steering the rewriting results. Furthermore, a fast verification pipeline is proposed to reduce verification cost. To cater to the needs of both industry and academia, we propose two benchmarking suites: the Large Rewriter Benchmark, targeting complex scenarios with extensive circuit partitioning, optimization trade-offs, and verification challenges, and the Small Rewriter Benchmark, designed for a wider range of scenarios and patterns. Our comparative analysis with established compilers such as Yosys and E-graph demonstrates significant improvements, highlighting the benefits of integrating large models into the early stages of circuit design. We provide our benchmarks at https://github.com/yaoxufeng/RTLRewriter-Bench., Comment: ICCAD2024

While previous audio-driven talking head generation (THG) methods generate head poses from driving audio, the generated poses or lips cannot match the audio well or are not editable. In this study, we propose \textbf{PoseTalk}, a THG system that can freely generate lip-synchronized talking head videos with free head poses conditioned on text prompts and audio. The core insight of our method is using head pose to connect visual, linguistic, and audio signals. First, we propose to generate poses from both audio and text prompts, where the audio offers short-term variations and rhythm correspondence of the head movements and the text prompts describe the long-term semantics of head motions. To achieve this goal, we devise a Pose Latent Diffusion (PLD) model to generate motion latent from text prompts and audio cues in a pose latent space. Second, we observe a loss-imbalance problem: the loss for the lip region contributes less than 4\% of the total reconstruction loss caused by both pose and lip, making optimization lean towards head movements rather than lip shapes. To address this issue, we propose a refinement-based learning strategy to synthesize natural talking videos using two cascaded networks, i.e., CoarseNet, and RefineNet. The CoarseNet estimates coarse motions to produce animated images in novel poses and the RefineNet focuses on learning finer lip motions by progressively estimating lip motions from low-to-high resolutions, yielding improved lip-synchronization performance. Experiments demonstrate our pose prediction strategy achieves better pose diversity and realness compared to text-only or audio-only, and our video generator model outperforms state-of-the-art methods in synthesizing talking videos with natural head motions. Project: https://junleen.github.io/projects/posetalk., Comment: 7+5 pages, 15 figures

The rapid growth of scientific literature imposes significant challenges for researchers endeavoring to stay updated with the latest advancements in their fields and delve into new areas. We introduce OpenResearcher, an innovative platform that leverages Artificial Intelligence (AI) techniques to accelerate the research process by answering diverse questions from researchers. OpenResearcher is built based on Retrieval-Augmented Generation (RAG) to integrate Large Language Models (LLMs) with up-to-date, domain-specific knowledge. Moreover, we develop various tools for OpenResearcher to understand researchers' queries, search from the scientific literature, filter retrieved information, provide accurate and comprehensive answers, and self-refine these answers. OpenResearcher can flexibly use these tools to balance efficiency and effectiveness. As a result, OpenResearcher enables researchers to save time and increase their potential to discover new insights and drive scientific breakthroughs. Demo, video, and code are available at: https://github.com/GAIR-NLP/OpenResearcher., Comment: Accepted to Demo track of EMNLP 2024

Rate splitting multiple access (RSMA) relies on beamforming design for attaining spectral efficiency and energy efficiency gains over traditional multiple access schemes. While conventional optimization approaches such as weighted minimum mean square error (WMMSE) achieve suboptimal solutions for RSMA beamforming optimization, they are computationally demanding. A novel approach based on fractional programming (FP) has unveiled the optimal beamforming structure (OBS) for RSMA. This method, combined with a hyperplane fixed point iteration (HFPI) approach, named FP-HFPI, provides suboptimal beamforming solutions with identical sum rate performance but much lower computational complexity compared to WMMSE. Inspired by such an approach, in this work, a novel deep unfolding framework based on FP-HFPI, named rate-splitting-beamforming neural network (RS-BNN), is proposed to unfold the FP-HFPI algorithm. Numerical results indicate that the proposed RS-BNN attains a level of performance closely matching that of WMMSE and FP-HFPI, while dramatically reducing the computational complexity.

Target sound extraction (TSE) separates the target sound from the mixture signals based on provided clues. However, the performance of existing models significantly degrades under reverberant conditions. Inspired by auditory scene analysis (ASA), this work proposes a TSE model provided with pitch information named TSE-PI. Conditional pitch extraction is achieved through the Feature-wise Linearly Modulated layer with the sound-class label. A modified Waveformer model combined with pitch information, employing a learnable Gammatone filterbank in place of the convolutional encoder, is used for target sound extraction. The inclusion of pitch information is aimed at improving the model's performance. The experimental results on the FSD50K dataset illustrate 2.4 dB improvements of target sound extraction under reverberant environments when incorporating pitch information and Gammatone filterbank., Comment: Accepted by Interspeech2024

Researchers have reported high decoding accuracy (>95%) using non-invasive Electroencephalogram (EEG) signals for brain-computer interface (BCI) decoding tasks like image decoding, emotion recognition, auditory spatial attention detection, etc. Since these EEG data were usually collected with well-designed paradigms in labs, the reliability and robustness of the corresponding decoding methods were doubted by some researchers, and they argued that such decoding accuracy was overestimated due to the inherent temporal autocorrelation of EEG signals. However, the coupling between the stimulus-driven neural responses and the EEG temporal autocorrelations makes it difficult to confirm whether this overestimation exists in truth. Furthermore, the underlying pitfalls behind overestimated decoding accuracy have not been fully explained due to a lack of appropriate formulation. In this work, we formulate the pitfall in various EEG decoding tasks in a unified framework. EEG data were recorded from watermelons to remove stimulus-driven neural responses. Labels were assigned to continuous EEG according to the experimental design for EEG recording of several typical datasets, and then the decoding methods were conducted. The results showed the label can be successfully decoded as long as continuous EEG data with the same label were split into training and test sets. Further analysis indicated that high accuracy of various BCI decoding tasks could be achieved by associating labels with EEG intrinsic temporal autocorrelation features. These results underscore the importance of choosing the right experimental designs and data splits in BCI decoding tasks to prevent inflated accuracies due to EEG temporal autocorrelation.

Colorizing grayscale images offers an engaging visual experience. Existing automatic colorization methods often fail to generate satisfactory results due to incorrect semantic colors and unsaturated colors. In this work, we propose an automatic colorization pipeline to overcome these challenges. We leverage the extraordinary generative ability of the diffusion prior to synthesize color with plausible semantics. To overcome the artifacts introduced by the diffusion prior, we apply the luminance conditional guidance. Moreover, we adopt multimodal high-level semantic priors to help the model understand the image content and deliver saturated colors. Besides, a luminance-aware decoder is designed to restore details and enhance overall visual quality. The proposed pipeline synthesizes saturated colors while maintaining plausible semantics. Experiments indicate that our proposed method considers both diversity and fidelity, surpassing previous methods in terms of perceptual realism and gain most human preference.

In this paper, we propose a class of codes, referred to as random staircase generator matrix codes (SGMCs), which have staircase-like generator matrices. In the infinite-length region, we prove that the random SGMC is capacity-achieving over binary-input output-symmetric (BIOS) channels. In the finite-length region, we present the representative ordered statistics decoding with local constraints (LC-ROSD) algorithm for the SGMCs. The most distinguished feature of the SGMCs with LC-ROSD is that the staircase-like matrices enable parallel implementation of the Gaussian elimination (GE), avoiding the serial GE of conventional OSD and supporting a potential low decoding latency, as implied from simulations. To analyze the performance of random SGMCs in the finite-length region, we derive the ensemble weight spectrum and invoke the conventional union bound. We also derive a partially random coding union (RCU) bound, which is tighter than the conventional one and is used as a criterion to design the SGMCs. Staircase-like generator matrices allow us to derive a series of (tighter and tighter) lower bounds based on the second-order Bonferroni inequality with the incremental number of codewords. The numerical results show that the decoding performance can match well with the proposed partially RCU bound for different code rates and different profiles. The numerical results also show that the tailored SGMCs with the LC-ROSD algorithm can approach the finite-length performance bound, outperforming the 5G low-density parity-check (LDPC) codes, 5G polar codes, and Reed-Muller (RM) codes.

The highly sensitive millimeter-wave telescope is an important tool for accurate measurement of Cosmic Microwave Background (CMB) radiation, and its core component is a detector array located in a cryogenic focal plane. The feasibility of utilizing thermal kinetic inductance detectors (TKIDs) for CMB observations has been demonstrated. We propose a pixel design of TKIDs for observing CMB through atmospheric windows for observations in the 90/150 GHz bands. Assuming lossless dielectric, the coupling efficiency of a single pixel is around 90%. This pixel design will be utilized for future large-scale TKIDs array designs for CMB observations.

It has been long that literature in financial academics focuses mainly on price and return but much less on trading volume. In the past twenty years, it has already linked both price and trading volume to economic fundamentals, and explored the behavioral implications of trading volume such as investor's attitude toward risks, overconfidence, disagreement, and attention etc. However, what is surprising is how little we really know about trading volume. Here we show that trading volume probability represents the frequency of market crowd's trading action in terms of behavior analysis, and test two adaptive hypotheses relevant to the volume uncertainty associated with price in China stock market. The empirical work reveals that market crowd trade a stock in efficient adaptation except for simple heuristics, gradually tend to achieve agreement on an outcome or an asset price widely on a trading day, and generate such a stationary equilibrium price very often in interaction and competition among themselves no matter whether it is highly overestimated or underestimated. This suggests that asset prices include not only a fundamental value but also private information, speculative, sentiment, attention, gamble, and entertainment values etc. Moreover, market crowd adapt to gain and loss by trading volume increase or decrease significantly in interaction with environment in any two consecutive trading days. Our results demonstrate how interaction between information and news, the trading action, and return outcomes in the three-term feedback loop produces excessive trading volume which includes various internal and external causes., Comment: 57 pages, 11 figures, 5 tables