Search

The communications and interrelations between different locations on the Earth's surface have far-reaching implications for both social and natural systems. Effective spatial analytics ideally require a spatial representation, where geographic principles are succinctly expressed within a defined metric space. However, common spatial representations, including map-based or network-based approaches, fall short by incompletely or inaccurately defining this metric space. Here we show, by introducing an inverse friction factor that captures the spatial constraints in spatial networks, that a homogeneous, low-dimensional spatial representation - termed the Geographic Manifold - can be achieved. We illustrate the effectiveness of the Geographic Manifold in two classic scenarios of spatial analytics - location choice and propagation, where the otherwise complicated analyses are reduced to straightforward regular partitioning and concentric diffusing, respectively on the manifold with a high degree of accuracy. We further empirically explain and formally prove the general existence of the Geographic Manifold, which is grounded in the intrinsic Euclidean low-dimensional statistical physics properties of geographic phenomena. This work represents a step towards formalizing Tobler's famous First Law of Geography from a geometric approach, where a regularized geospace thereby yielded is expected to contribute in learning abstract spatial structure representations for understanding and optimization purposes.

Current methods of building LLMs with voice interaction capabilities rely heavily on explicit text autoregressive generation before or during speech response generation to maintain content quality, which unfortunately brings computational overhead and increases latency in multi-turn interactions. To address this, we introduce IntrinsicVoic,e an LLM designed with intrinsic real-time voice interaction capabilities. IntrinsicVoice aims to facilitate the transfer of textual capabilities of pre-trained LLMs to the speech modality by mitigating the modality gap between text and speech. Our novelty architecture, GroupFormer, can reduce speech sequences to lengths comparable to text sequences while generating high-quality audio, significantly reducing the length difference between speech and text, speeding up inference, and alleviating long-text modeling issues. Additionally, we construct a multi-turn speech-to-speech dialogue dataset named \method-500k which includes nearly 500k turns of speech-to-speech dialogues, and a cross-modality training strategy to enhance the semantic alignment between speech and text. Experimental results demonstrate that IntrinsicVoice can generate high-quality speech response with latency lower than 100ms in multi-turn dialogue scenarios. Demos are available at https://instrinsicvoice.github.io/.

Time series forecasting is prevalent in extensive real-world applications, such as financial analysis and energy planning. Previous studies primarily focus on time series modality, endeavoring to capture the intricate variations and dependencies inherent in time series. Beyond numerical time series data, we notice that metadata (e.g.~dataset and variate descriptions) also carries valuable information essential for forecasting, which can be used to identify the application scenario and provide more interpretable knowledge than digit sequences. Inspired by this observation, we propose a Metadata-informed Time Series Transformer (MetaTST), which incorporates multiple levels of context-specific metadata into Transformer forecasting models to enable informative time series forecasting. To tackle the unstructured nature of metadata, MetaTST formalizes them into natural languages by pre-designed templates and leverages large language models (LLMs) to encode these texts into metadata tokens as a supplement to classic series tokens, resulting in an informative embedding. Further, a Transformer encoder is employed to communicate series and metadata tokens, which can extend series representations by metadata information for more accurate forecasting. This design also allows the model to adaptively learn context-specific patterns across various scenarios, which is particularly effective in handling large-scale, diverse-scenario forecasting tasks. Experimentally, MetaTST achieves state-of-the-art compared to advanced time series models and LLM-based methods on widely acknowledged short- and long-term forecasting benchmarks, covering both single-dataset individual and multi-dataset joint training settings.

We introduce Seed-Music, a suite of music generation systems capable of producing high-quality music with fine-grained style control. Our unified framework leverages both auto-regressive language modeling and diffusion approaches to support two key music creation workflows: controlled music generation and post-production editing. For controlled music generation, our system enables vocal music generation with performance controls from multi-modal inputs, including style descriptions, audio references, musical scores, and voice prompts. For post-production editing, it offers interactive tools for editing lyrics and vocal melodies directly in the generated audio. We encourage readers to listen to demo audio examples at https://team.doubao.com/seed-music "https://team.doubao.com/seed-music"., Comment: Seed-Music technical report, 20 pages, 5 figures

Speech self-supervised pre-training can effectively improve the performance of downstream tasks. However, previous self-supervised learning (SSL) methods for speech, such as HuBERT and BEST-RQ, focus on utilizing non-causal encoders with bidirectional context, and lack sufficient support for downstream streaming models. To address this issue, we introduce the next token prediction based speech pre-training method with random-projection quantizer (NEST-RQ). NEST-RQ employs causal encoders with only left context and uses next token prediction (NTP) as the training task. On the large-scale dataset, compared to BEST-RQ, the proposed NEST-RQ achieves comparable performance on non-streaming automatic speech recognition (ASR) and better performance on streaming ASR. We also conduct analytical experiments in terms of the future context size of streaming ASR, the codebook quality of SSL and the model size of the encoder. In summary, the paper demonstrates the feasibility of the NTP in speech SSL and provides empirical evidence and insights for speech SSL research., Comment: 5 pages, 2 figures, Work in progress

We obtain the upper error bounds of robust estimators for mean vector, using the median-of-means (MOM) method. The method is designed to handle data with heavy tails and contamination, with only a finite second moment, which is weaker than many others, relying on the VC dimension rather than the Rademacher complexity to measure statistical complexity. This allows us to implement MOM in covariance estimation, without imposing conditions such as $L$-sub-Gaussian or $L_{4}-L_{2}$ norm equivalence. In particular, we derive a new robust estimator, the MOM version of the halfspace depth, along with error bounds for mean estimation in any norm.

Recent advancements in large-scale video-language models have shown significant potential for real-time planning and detailed interactions. However, their high computational demands and the scarcity of annotated datasets limit their practicality for academic researchers. In this work, we introduce VideoLLaMB, a novel framework that utilizes temporal memory tokens within bridge layers to allow for the encoding of entire video sequences alongside historical visual data, effectively preserving semantic continuity and enhancing model performance across various tasks. This approach includes recurrent memory tokens and a SceneTilling algorithm, which segments videos into independent semantic units to preserve semantic integrity. Empirically, VideoLLaMB significantly outstrips existing video-language models, demonstrating a 5.5 points improvement over its competitors across three VideoQA benchmarks, and 2.06 points on egocentric planning. Comprehensive results on the MVBench show that VideoLLaMB-7B achieves markedly better results than previous 7B models of same LLM. Remarkably, it maintains robust performance as PLLaVA even as video length increases up to 8 times. Besides, the frame retrieval results on our specialized Needle in a Video Haystack (NIAVH) benchmark, further validate VideoLLaMB's prowess in accurately identifying specific frames within lengthy videos. Our SceneTilling algorithm also enables the generation of streaming video captions directly, without necessitating additional training. In terms of efficiency, VideoLLaMB, trained on 16 frames, supports up to 320 frames on a single Nvidia A100 GPU with linear GPU memory scaling, ensuring both high performance and cost-effectiveness, thereby setting a new foundation for long-form video-language models in both academic and practical applications.

Hallucination is a common issue in Multimodal Large Language Models (MLLMs), yet the underlying principles remain poorly understood. In this paper, we investigate which components of MLLMs contribute to object hallucinations. To analyze image representations while completely avoiding the influence of all other factors other than the image representation itself, we propose a parametric-free representation alignment metric (Pfram) that can measure the similarities between any two representation systems without requiring additional training parameters. Notably, Pfram can also assess the alignment of a neural representation system with the human representation system, represented by ground-truth annotations of images. By evaluating the alignment with object annotations, we demonstrate that this metric shows strong and consistent correlations with object hallucination across a wide range of state-of-the-art MLLMs, spanning various model architectures and sizes. Furthermore, using this metric, we explore other key issues related to image representations in MLLMs, such as the role of different modules, the impact of textual instructions, and potential improvements including the use of alternative visual encoders. Our code is available at: https://github.com/yellow-binary-tree/Pfram.

We present a bosonized effective field theory for a 2d Fermi surface in a weak magnetic field using the coadjoint orbit approach, which was recently developed as a nonlinear bosonization method in phase space for Fermi liquids and non-Fermi liquids. We show that by parametrizing the phase space with the guiding center and the mechanical momentum, and by using techniques in noncommutative field theory, the physics of Landau levels and Landau level degeneracy ($N_{\Phi}$) naturally arises. For a parabolic dispersion, the resulting theory describes $N_{\Phi}$ flavors of free chiral bosons propagating in \emph{momentum space}. In addition, the action contains a linear term in the bosonic field, which upon mode expansion becomes a topological $\theta$-term. By properly quantizing this theory, we reproduce the well-known thermal and magnetic responses of a Fermi surface, including linear-in-$T$ specific heat, Landau diamagnetism, and the de Haas-van Alphen effect. {In particular, the de Haas-van Alphen effect is shown to be a direct consequence of the topological $\theta$-term.} Our theory paves the way toward understanding correlated gapless fermionic systems in a magnetic field using the powerful approach of bosonization., Comment: 16 pages, 2 figures

Dialogue serves as the most natural manner of human-computer interaction (HCI). Recent advancements in speech language models (SLM) have significantly enhanced speech-based conversational AI. However, these models are limited to turn-based conversation, lacking the ability to interact with humans in real-time spoken scenarios, for example, being interrupted when the generated content is not satisfactory. To address these limitations, we explore full duplex modeling (FDM) in interactive speech language models (iSLM), focusing on enhancing real-time interaction and, more explicitly, exploring the quintessential ability of interruption. We introduce a novel model design, namely listening-while-speaking language model (LSLM), an end-to-end system equipped with both listening and speaking channels. Our LSLM employs a token-based decoder-only TTS for speech generation and a streaming self-supervised learning (SSL) encoder for real-time audio input. LSLM fuses both channels for autoregressive generation and detects turn-taking in real time. Three fusion strategies -- early fusion, middle fusion, and late fusion -- are explored, with middle fusion achieving an optimal balance between speech generation and real-time interaction. Two experimental settings, command-based FDM and voice-based FDM, demonstrate LSLM's robustness to noise and sensitivity to diverse instructions. Our results highlight LSLM's capability to achieve duplex communication with minimal impact on existing systems. This study aims to advance the development of interactive speech dialogue systems, enhancing their applicability in real-world contexts., Comment: Demo can be found at https://ddlbojack.github.io/LSLM

Compositional visual reasoning methods, which translate a complex query into a structured composition of feasible visual tasks, have exhibited a strong potential in complicated multi-modal tasks. Empowered by recent advances in large language models (LLMs), this multi-modal challenge has been brought to a new stage by treating LLMs as few-shot/zero-shot planners, i.e., vision-language (VL) programming. Such methods, despite their numerous merits, suffer from challenges due to LLM planning mistakes or inaccuracy of visual execution modules, lagging behind the non-compositional models. In this work, we devise a "plug-and-play" method, ExoViP, to correct errors in both the planning and execution stages through introspective verification. We employ verification modules as "exoskeletons" to enhance current VL programming schemes. Specifically, our proposed verification module utilizes a mixture of three sub-verifiers to validate predictions after each reasoning step, subsequently calibrating the visual module predictions and refining the reasoning trace planned by LLMs. Experimental results on two representative VL programming methods showcase consistent improvements on five compositional reasoning tasks on standard benchmarks. In light of this, we believe that ExoViP can foster better performance and generalization on open-domain multi-modal challenges., Comment: To Appear at COLM 2024

Time series, characterized by a sequence of data points arranged in a discrete-time order, are ubiquitous in real-world applications. Different from other modalities, time series present unique challenges due to their complex and dynamic nature, including the entanglement of nonlinear patterns and time-variant trends. Analyzing time series data is of great significance in real-world scenarios and has been widely studied over centuries. Recent years have witnessed remarkable breakthroughs in the time series community, with techniques shifting from traditional statistical methods to advanced deep learning models. In this paper, we delve into the design of deep time series models across various analysis tasks and review the existing literature from two perspectives: basic modules and model architectures. Further, we develop and release Time Series Library (TSLib) as a fair benchmark of deep time series models for diverse analysis tasks, which implements 24 mainstream models, covers 30 datasets from different domains, and supports five prevalent analysis tasks. Based on TSLib, we thoroughly evaluate 12 advanced deep time series models on different tasks. Empirical results indicate that models with specific structures are well-suited for distinct analytical tasks, which offers insights for research and adoption of deep time series models. Code is available at https://github.com/thuml/Time-Series-Library., Comment: \

Modern automatic speech recognition (ASR) model is required to accurately transcribe diverse speech signals (from different domains, languages, accents, etc) given the specific contextual information in various application scenarios. Classic end-to-end models fused with extra language models perform well, but mainly in data matching scenarios and are gradually approaching a bottleneck. In this work, we introduce Seed-ASR, a large language model (LLM) based speech recognition model. Seed-ASR is developed based on the framework of audio conditioned LLM (AcLLM), leveraging the capabilities of LLMs by inputting continuous speech representations together with contextual information into the LLM. Through stage-wise large-scale training and the elicitation of context-aware capabilities in LLM, Seed-ASR demonstrates significant improvement over end-to-end models on comprehensive evaluation sets, including multiple domains, accents/dialects and languages. Additionally, Seed-ASR can be further deployed to support specific needs in various scenarios without requiring extra language models. Compared to recently released large ASR models, Seed-ASR achieves 10%-40% reduction in word (or character, for Chinese) error rates on Chinese and English public test sets, further demonstrating its powerful performance.

Generative models have shown significant achievements in audio generation tasks. However, existing models struggle with complex and detailed prompts, leading to potential performance degradation. We hypothesize that this problem stems from the simplicity and scarcity of the training data. This work aims to create a large-scale audio dataset with rich captions for improving audio generation models. We first develop an automated pipeline to generate detailed captions by transforming predicted visual captions, audio captions, and tagging labels into comprehensive descriptions using a Large Language Model (LLM). The resulting dataset, Sound-VECaps, comprises 1.66M high-quality audio-caption pairs with enriched details including audio event orders, occurred places and environment information. We then demonstrate that training the text-to-audio generation models with Sound-VECaps significantly improves the performance on complex prompts. Furthermore, we conduct ablation studies of the models on several downstream audio-language tasks, showing the potential of Sound-VECaps in advancing audio-text representation learning. Our dataset and models are available online., Comment: 5 pages with 1 appendix

This report introduces FunAudioLLM, a model family designed to enhance natural voice interactions between humans and large language models (LLMs). At its core are two innovative models: SenseVoice, which handles multilingual speech recognition, emotion recognition, and audio event detection; and CosyVoice, which facilitates natural speech generation with control over multiple languages, timbre, speaking style, and speaker identity. SenseVoice-Small delivers exceptionally low-latency ASR for 5 languages, and SenseVoice-Large supports high-precision ASR for over 50 languages, while CosyVoice excels in multi-lingual voice generation, zero-shot in-context learning, cross-lingual voice cloning, and instruction-following capabilities. The models related to SenseVoice and CosyVoice have been open-sourced on Modelscope and Huggingface, along with the corresponding training, inference, and fine-tuning codes released on GitHub. By integrating these models with LLMs, FunAudioLLM enables applications such as speech-to-speech translation, emotional voice chat, interactive podcasts, and expressive audiobook narration, thereby pushing the boundaries of voice interaction technology. Demos are available at https://fun-audio-llm.github.io, and the code can be accessed at https://github.com/FunAudioLLM., Comment: Work in progress. Authors are listed in alphabetical order by family name

Despite the rapid development of Chinese vision-language models (VLMs), most existing Chinese vision-language (VL) datasets are constructed on Western-centric images from existing English VL datasets. The cultural bias in the images makes these datasets unsuitable for evaluating VLMs in Chinese culture. To remedy this issue, we present a new Chinese Vision- Language Understanding Evaluation (CVLUE) benchmark dataset, where the selection of object categories and images is entirely driven by Chinese native speakers, ensuring that the source images are representative of Chinese culture. The benchmark contains four distinct VL tasks ranging from image-text retrieval to visual question answering, visual grounding and visual dialogue. We present a detailed statistical analysis of CVLUE and provide a baseline performance analysis with several open-source multilingual VLMs on CVLUE and its English counterparts to reveal their performance gap between English and Chinese. Our in-depth category-level analysis reveals a lack of Chinese cultural knowledge in existing VLMs. We also find that fine-tuning on Chinese culture-related VL datasets effectively enhances VLMs' understanding of Chinese culture.

Recent works have shown promising results in connecting speech encoders to large language models (LLMs) for speech recognition. However, several limitations persist, including limited fine-tuning options, a lack of mechanisms to enforce speech-text alignment, and high insertion errors especially in domain mismatch conditions. This paper presents a comprehensive solution to address these issues. We begin by investigating more thoughtful fine-tuning schemes. Next, we propose a matching loss to enhance alignment between modalities. Finally, we explore training and inference methods to mitigate high insertion errors. Experimental results on the Librispeech corpus demonstrate that partially fine-tuning the encoder and LLM using parameter-efficient methods, such as LoRA, is the most cost-effective approach. Additionally, the matching loss improves modality alignment, enhancing performance. The proposed training and inference methods significantly reduce insertion errors.

Recent advancements in Multimodal Large Language Models (MLLMs) have extended their capabilities to video understanding. Yet, these models are often plagued by "hallucinations", where irrelevant or nonsensical content is generated, deviating from the actual video context. This work introduces VideoHallucer, the first comprehensive benchmark for hallucination detection in large video-language models (LVLMs). VideoHallucer categorizes hallucinations into two main types: intrinsic and extrinsic, offering further subcategories for detailed analysis, including object-relation, temporal, semantic detail, extrinsic factual, and extrinsic non-factual hallucinations. We adopt an adversarial binary VideoQA method for comprehensive evaluation, where pairs of basic and hallucinated questions are crafted strategically. By evaluating eleven LVLMs on VideoHallucer, we reveal that i) the majority of current models exhibit significant issues with hallucinations; ii) while scaling datasets and parameters improves models' ability to detect basic visual cues and counterfactuals, it provides limited benefit for detecting extrinsic factual hallucinations; iii) existing models are more adept at detecting facts than identifying hallucinations. As a byproduct, these analyses further instruct the development of our self-PEP framework, achieving an average of 5.38% improvement in hallucination resistance across all model architectures.

Speech understanding as an element of the more generic video understanding using audio-visual large language models (av-LLMs) is a crucial yet understudied aspect. This paper proposes video-SALMONN, a single end-to-end av-LLM for video processing, which can understand not only visual frame sequences, audio events and music, but speech as well. To obtain fine-grained temporal information required by speech understanding, while keeping efficient for other video elements, this paper proposes a novel multi-resolution causal Q-Former (MRC Q-Former) structure to connect pre-trained audio-visual encoders and the backbone large language model. Moreover, dedicated training approaches including the diversity loss and the unpaired audio-visual mixed training scheme are proposed to avoid frames or modality dominance. On the introduced speech-audio-visual evaluation benchmark, video-SALMONN achieves more than 25\% absolute accuracy improvements on the video-QA task and over 30\% absolute accuracy improvements on audio-visual QA tasks with human speech. In addition, video-SALMONN demonstrates remarkable video comprehension and reasoning abilities on tasks that are unprecedented by other av-LLMs. Our training code and model checkpoints are available at \texttt{\url{https://github.com/bytedance/SALMONN/}}., Comment: Accepted at ICML 2024. arXiv admin note: substantial text overlap with arXiv:2310.05863

Speech encompasses a wealth of information, including but not limited to content, paralinguistic, and environmental information. This comprehensive nature of speech significantly impacts communication and is crucial for human-computer interaction. Chat-Oriented Large Language Models (LLMs), known for their general-purpose assistance capabilities, have evolved to handle multi-modal inputs, including speech. Although these models can be adept at recognizing and analyzing speech, they often fall short of generating appropriate responses. We argue that this is due to the lack of principles on task definition and model development, which requires open-source datasets and metrics suitable for model evaluation. To bridge the gap, we present SD-Eval, a benchmark dataset aimed at multidimensional evaluation of spoken dialogue understanding and generation. SD-Eval focuses on paralinguistic and environmental information and includes 7,303 utterances, amounting to 8.76 hours of speech data. The data is aggregated from eight public datasets, representing four perspectives: emotion, accent, age, and background sound. To assess the SD-Eval benchmark dataset, we implement three different models and construct a training set following a similar process as SD-Eval. The training set contains 1,052.72 hours of speech data and 724.4k utterances. We also conduct a comprehensive evaluation using objective evaluation methods (e.g. BLEU and ROUGE), subjective evaluations and LLM-based metrics for the generated responses. Models conditioned with paralinguistic and environmental information outperform their counterparts in both objective and subjective measures. Moreover, experiments demonstrate LLM-based metrics show a higher correlation with human evaluation compared to traditional metrics. We open-source SD-Eval at https://github.com/amphionspace/SD-Eval.

Causal discovery is crucial for understanding complex systems and informing decisions. While observational data can uncover causal relationships under certain assumptions, it often falls short, making active interventions necessary. Current methods, such as Bayesian and graph-theoretical approaches, do not prioritize decision-making and often rely on ideal conditions or information gain, which is not directly related to hypothesis testing. We propose a novel Bayesian optimization-based method inspired by Bayes factors that aims to maximize the probability of obtaining decisive and correct evidence. Our approach uses observational data to estimate causal models under different hypotheses, evaluates potential interventions pre-experimentally, and iteratively updates priors to refine interventions. We demonstrate the effectiveness of our method through various experiments. Our contributions provide a robust framework for efficient causal discovery through active interventions, enhancing the practical application of theoretical advancements.

This paper explores enabling large language models (LLMs) to understand spatial information from multichannel audio, a skill currently lacking in auditory LLMs. By leveraging LLMs' advanced cognitive and inferential abilities, the aim is to enhance understanding of 3D environments via audio. We study 3 spatial audio tasks: sound source localization (SSL), far-field speech recognition (FSR), and localisation-informed speech extraction (LSE), achieving notable progress in each task. For SSL, our approach achieves an MAE of $2.70^{\circ}$ on the Spatial LibriSpeech dataset, substantially surpassing the prior benchmark of about $6.60^{\circ}$. Moreover, our model can employ spatial cues to improve FSR accuracy and execute LSE by selectively attending to sounds originating from a specified direction via text prompts, even amidst overlapping speech. These findings highlight the potential of adapting LLMs to grasp physical audio concepts, paving the way for LLM-based agents in 3D environments., Comment: Accepted at Interspeech 2024

This paper addresses challenges in integrating new languages into a pre-trained multilingual automatic speech recognition (mASR) system, particularly in scenarios where training data for existing languages is limited or unavailable. The proposed method employs a dual-pipeline with low-rank adaptation (LoRA). It maintains two data flow pipelines-one for existing languages and another for new languages. The primary pipeline follows the standard flow through the pre-trained parameters of mASR, while the secondary pipeline additionally utilizes language-specific parameters represented by LoRA and a separate output decoder module. Importantly, the proposed approach minimizes the performance degradation of existing languages and enables a language-agnostic operation mode, facilitated by a decoder selection strategy. We validate the effectiveness of the proposed method by extending the pre-trained Whisper model to 19 new languages from the FLEURS dataset, Comment: 5 pages, 2 figures, 4 tables

We introduce Seed-TTS, a family of large-scale autoregressive text-to-speech (TTS) models capable of generating speech that is virtually indistinguishable from human speech. Seed-TTS serves as a foundation model for speech generation and excels in speech in-context learning, achieving performance in speaker similarity and naturalness that matches ground truth human speech in both objective and subjective evaluations. With fine-tuning, we achieve even higher subjective scores across these metrics. Seed-TTS offers superior controllability over various speech attributes such as emotion and is capable of generating highly expressive and diverse speech for speakers in the wild. Furthermore, we propose a self-distillation method for speech factorization, as well as a reinforcement learning approach to enhance model robustness, speaker similarity, and controllability. We additionally present a non-autoregressive (NAR) variant of the Seed-TTS model, named $\text{Seed-TTS}_\text{DiT}$, which utilizes a fully diffusion-based architecture. Unlike previous NAR-based TTS systems, $\text{Seed-TTS}_\text{DiT}$ does not depend on pre-estimated phoneme durations and performs speech generation through end-to-end processing. We demonstrate that this variant achieves comparable performance to the language model-based variant and showcase its effectiveness in speech editing. We encourage readers to listen to demos at \url{https://bytedancespeech.github.io/seedtts_tech_report}.

Audio and visual signals typically occur simultaneously, and humans possess an innate ability to correlate and synchronize information from these two modalities. Recently, a challenging problem known as Audio-Visual Segmentation (AVS) has emerged, intending to produce segmentation maps for sounding objects within a scene. However, the methods proposed so far have not sufficiently integrated audio and visual information, and the computational costs have been extremely high. Additionally, the outputs of different stages have not been fully utilized. To facilitate this research, we introduce a novel Progressive Confident Masking Attention Network (PMCANet). It leverages attention mechanisms to uncover the intrinsic correlations between audio signals and visual frames. Furthermore, we design an efficient and effective cross-attention module to enhance semantic perception by selecting query tokens. This selection is determined through confidence-driven units based on the network's multi-stage predictive outputs. Experiments demonstrate that our network outperforms other AVS methods while requiring less computational resources., Comment: 10 pages, 9 figures, submitted to IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY

The Stable Diffusion Model (SDM) is a prevalent and effective model for text-to-image (T2I) and image-to-image (I2I) generation. Despite various attempts at sampler optimization, model distillation, and network quantification, these approaches typically maintain the original network architecture. The extensive parameter scale and substantial computational demands have limited research into adjusting the model architecture. This study focuses on reducing redundant computation in SDM and optimizes the model through both tuning and tuning-free methods. 1) For the tuning method, we design a model assembly strategy to reconstruct a lightweight model while preserving performance through distillation. Second, to mitigate performance loss due to pruning, we incorporate multi-expert conditional convolution (ME-CondConv) into compressed UNets to enhance network performance by increasing capacity without sacrificing speed. Third, we validate the effectiveness of the multi-UNet switching method for improving network speed. 2) For the tuning-free method, we propose a feature inheritance strategy to accelerate inference by skipping local computations at the block, layer, or unit level within the network structure. We also examine multiple sampling modes for feature inheritance at the time-step level. Experiments demonstrate that both the proposed tuning and the tuning-free methods can improve the speed and performance of the SDM. The lightweight model reconstructed by the model assembly strategy increases generation speed by $22.4%$, while the feature inheritance strategy enhances the SDM generation speed by $40.0%$., Comment: 19 pages, 16 figures, submitted to IEEE Transactions on Neural Networks and Learning Systems

Medium and high entropy ceramics, with their distinctive disordered structures, exhibit ultra-low thermal conductivity and high temperature stability. These properties make them strong contenders for next generation thermal barrier coating (TBC) materials. However, predicting their thermal conductivity has been challenging, primarily due to their unique phonon scattering mechanisms. Apart from the conventional phonon-phonon scattering mechanism, the phonon-disorder scattering, comprising both mass and force constant disorder, are also expected to make significant contribution in determining the thermal conductivity of medium and high entropy ceramics. However, it remains challenging to quantify the phonon-disorder contribution, particular in the aspect of force constant disorder. Here we demonstrated a relationship between the lattice disorder, a quantity more readily calculable, with force constant disorder, rendering it possible to substitute the force constant disorder by lattice disorder. Based on this relationship and drawing inspiration from Klement's equation of static imperfection, we have developed a model that quantitatively assesses the connection between disorder and thermal conductivity. Applying our model to the medium/high entropy rocksalt and pyrochlore oxides as representatives, we found good alignment between the theoretical predictions and experimental measurements of thermal conductivities, confirming the validity of our model. The model developed offers a critical predictive tool for rapid screening of TBC materials based on medium and high entropy ceramics., Comment: 25 pages, 7 figures

Large Language Model (LLM) -in-the-loop applications have been shown to effectively interpret the human user's commands, make plans, and operate external tools/systems accordingly. Still, the operation scope of the LLM agent is limited to passively following the user, requiring the user to frame his/her needs with regard to the underlying tools/systems. We note that the potential of an LLM-Agent User Interface (LAUI) is much greater. A user mostly ignorant to the underlying tools/systems should be able to work with a LAUI to discover an emergent workflow. Contrary to the conventional way of designing an explorable GUI to teach the user a predefined set of ways to use the system, in the ideal LAUI, the LLM agent is initialized to be proficient with the system, proactively studies the user and his/her needs, and proposes new interaction schemes to the user. To illustrate LAUI, we present Flute X GPT, a concrete example using an LLM agent, a prompt manager, and a flute-tutoring multi-modal software-hardware system to facilitate the complex, real-time user experience of learning to play the flute.

This paper surveys and organizes research works on medical dialog systems, which is an important yet challenging task. Although these systems have been surveyed in the medical community from an application perspective, a systematic review from a rigorous technical perspective has to date remained noticeably absent. As a result, an overview of the categories, methods, and evaluation of medical dialogue systems remain limited and underspecified, hindering the further improvement of this area. To fill this gap, we investigate an initial pool of 325 papers from well-known computer science, and natural language processing conferences and journals, and make an overview. Recently, large language models have shown strong model capacity on downstream tasks, which also reshaped medical dialog systems' foundation. Despite the alluring practical application value, current medical dialogue systems still suffer from problems. To this end, this paper lists the grand challenges of medical dialog systems, especially of large language models.

The Partitioning Min-Max Weighted Matching (PMMWM) problem, being a practical NP-hard problem, integrates the task of partitioning the vertices of a bipartite graph into disjoint sets of limited size with the classical Maximum-Weight Perfect Matching (MPWM) problem. Initially introduced in 2015, the state-of-the-art method for addressing PMMWM is the MP$_{\text{LS}}$. In this paper, we present a novel approach, the Fast Iterative Match-Partition Hybrid Genetic Algorithm (FIMP-HGA), for addressing PMMWM. Similar to MP$_{\text{LS}}$, FIMP-HGA divides the solving into match and partition stages, iteratively refining the solution. In the match stage, we propose the KM-M algorithm, which reduces matching complexity through incremental adjustments, significantly enhancing runtime efficiency. For the partition stage, we introduce a Hybrid Genetic Algorithm (HGA) incorporating an elite strategy and design a Greedy Partition Crossover (GPX) operator alongside a Multilevel Local Search (MLS) to optimize individuals in the population. Population initialization employs various methods, including the multi-way Karmarkar-Karp (KK) algorithm, ensuring both quality and diversity. At each iteration, the bipartite graph is adjusted based on the current solution, aiming for continuous improvement. To conduct comprehensive experiments, we develop a new instance generation method compatible with existing approaches, resulting in four benchmark groups. Extensive experiments evaluate various algorithm modules, accurately assessing each module's impact on improvement. Evaluation results on our benchmarks demonstrate that the proposed FIMP-HGA significantly enhances solution quality compared to MP$_{\text{LS}}$, meanwhile reducing runtime by 3 to 20 times.

Environment awareness is crucial for enhancing walking safety and stability of amputee wearing powered prosthesis when crossing uneven terrains such as stairs and obstacles. However, existing environmental perception systems for prosthesis only provide terrain types and corresponding parameters, which fails to prevent potential collisions when crossing uneven terrains and may lead to falls and other severe consequences. In this paper, a visual-inertial motion estimation approach is proposed for prosthesis to perceive its movement and the changes of spatial relationship between the prosthesis and uneven terrain when traversing them. To achieve this, we estimate the knee motion by utilizing a depth camera to perceive the environment and align feature points extracted from stairs and obstacles. Subsequently, an error-state Kalman filter is incorporated to fuse the inertial data into visual estimations to reduce the feature extraction error and obtain a more robust estimation. The motion of prosthetic joint and toe are derived using the prosthesis model parameters. Experiment conducted on our collected dataset and stair walking trials with a powered prosthesis shows that the proposed method can accurately tracking the motion of the human leg and prosthesis with an average root-mean-square error of toe trajectory less than 5 cm. The proposed method is expected to enable the environmental adaptive control for prosthesis, thereby enhancing amputee's safety and mobility in uneven terrains.

The signing key exposure of Certificate Authorities (CAs) remains a critical concern in PKI. These keys can be exposed by carefully designed attacks or operational errors even today. Traditional protections fail to eliminate such risk and one leaked key is enough to compromise the CA. This long-standing dilemma motivates us to consider removing CAs' signing keys and propose Armored Core, the first PKI security extension using the trusted binding of Physically Unclonable Function (PUF) for certificate operations. It makes key exposure impossible by eliminating the digital signing keys in CA. To achieve this, we design a set of PUF-based X.509v3 certificate functions for CAs to generate physically trusted "signatures" without using a digital key. Moreover, we introduce a novel PUF transparency mechanism to effectively monitor the PUF operations in CAs. We integrate Armored Core into real-world PKI systems including Let's Encrypt Pebble and Certbot. We also provide a PUF-embedded RISC-V CPU prototype. The evaluation results show that Armored Core can offer stronger security guarantees through signing key removal and without causing any extra overhead, but improves the overall performance by 11% on storage and 4.9%-73.7% on computation., Comment: Under peer review

We present VoiceShop, a novel speech-to-speech framework that can modify multiple attributes of speech, such as age, gender, accent, and speech style, in a single forward pass while preserving the input speaker's timbre. Previous works have been constrained to specialized models that can only edit these attributes individually and suffer from the following pitfalls: the magnitude of the conversion effect is weak, there is no zero-shot capability for out-of-distribution speakers, or the synthesized outputs exhibit undesirable timbre leakage. Our work proposes solutions for each of these issues in a simple modular framework based on a conditional diffusion backbone model with optional normalizing flow-based and sequence-to-sequence speaker attribute-editing modules, whose components can be combined or removed during inference to meet a wide array of tasks without additional model finetuning. Audio samples are available at \url{https://voiceshopai.github.io}.

Scene Graph Generation (SGG) provides basic language representation of visual scenes, requiring models to grasp complex and diverse semantics between various objects. However, this complexity and diversity in SGG also leads to underrepresentation, where part of test triplets are rare or even unseen during training, resulting in imprecise predictions. To tackle this, we propose using the SGG models with pretrained vision-language models (VLMs) to enhance representation. However, due to the gap between the pretraining and SGG, directly ensembling the pretrained VLMs leads to severe biases across relation words. Thus, we introduce LM Estimation to approximate the words' distribution underlies in the pretraining language sets, and then use the distribution for debiasing. After that, we ensemble VLMs with SGG models to enhance representation. Considering that each model may represent better at different samples, we use a certainty-aware indicator to score each sample and dynamically adjust the ensemble weights. Our method effectively addresses the words biases, enhances SGG's representation, and achieve markable performance enhancements. It is training-free and integrates well with existing SGG models.

The advent of 3D Gaussian Splatting (3DGS) has revolutionized 3D editing, offering efficient, high-fidelity rendering and enabling precise local manipulations. Currently, diffusion-based 2D editing models are harnessed to modify multi-view rendered images, which then guide the editing of 3DGS models. However, this approach faces a critical issue of multi-view inconsistency, where the guidance images exhibit significant discrepancies across views, leading to mode collapse and visual artifacts of 3DGS. To this end, we introduce View-consistent Editing (VcEdit), a novel framework that seamlessly incorporates 3DGS into image editing processes, ensuring multi-view consistency in edited guidance images and effectively mitigating mode collapse issues. VcEdit employs two innovative consistency modules: the Cross-attention Consistency Module and the Editing Consistency Module, both designed to reduce inconsistencies in edited images. By incorporating these consistency modules into an iterative pattern, VcEdit proficiently resolves the issue of multi-view inconsistency, facilitating high-quality 3DGS editing across a diverse range of scenes. Further video results are shown in http://vcedit.github.io., Comment: accepted to ECCV 2024

Video-text Large Language Models (video-text LLMs) have shown remarkable performance in answering questions and holding conversations on simple videos. However, they perform almost the same as random on grounding text queries in long and complicated videos, having little ability to understand and reason about temporal information, which is the most fundamental difference between videos and images. In this paper, we propose HawkEye, one of the first video-text LLMs that can perform temporal video grounding in a fully text-to-text manner. To collect training data that is applicable for temporal video grounding, we construct InternVid-G, a large-scale video-text corpus with segment-level captions and negative spans, with which we introduce two new time-aware training objectives to video-text LLMs. We also propose a coarse-grained method of representing segments in videos, which is more robust and easier for LLMs to learn and follow than other alternatives. Extensive experiments show that HawkEye is better at temporal video grounding and comparable on other video-text tasks with existing video-text LLMs, which verifies its superior video-text multi-modal understanding abilities.

Deep models have demonstrated remarkable performance in time series forecasting. However, due to the partially-observed nature of real-world applications, solely focusing on the target of interest, so-called endogenous variables, is usually insufficient to guarantee accurate forecasting. Notably, a system is often recorded into multiple variables, where the exogenous variables can provide valuable external information for endogenous variables. Thus, unlike well-established multivariate or univariate forecasting paradigms that either treat all the variables equally or ignore exogenous information, this paper focuses on a more practical setting: time series forecasting with exogenous variables. We propose a novel approach, TimeXer, to ingest external information to enhance the forecasting of endogenous variables. With deftly designed embedding layers, TimeXer empowers the canonical Transformer with the ability to reconcile endogenous and exogenous information, where patch-wise self-attention and variate-wise cross-attention are used simultaneously. Moreover, global endogenous tokens are learned to effectively bridge the causal information underlying exogenous series into endogenous temporal patches. Experimentally, TimeXer achieves consistent state-of-the-art performance on twelve real-world forecasting benchmarks and exhibits notable generality and scalability. Code is available at this repository: https://github.com/thuml/TimeXer.

Despite progress in multimodal large language models (MLLMs), the challenge of interpreting long-form videos in response to linguistic queries persists, largely due to the inefficiency in temporal grounding and limited pre-trained context window size. In this work, we introduce Temporal Grounding Bridge (TGB), a novel framework that bootstraps MLLMs with advanced temporal grounding capabilities and broadens their contextual scope. Our framework significantly enhances the temporal capabilities of current MLLMs through three key innovations: an efficient multi-span temporal grounding algorithm applied to low-dimension temporal features projected from flow; a multimodal length extrapolation training paradigm that utilizes low-dimension temporal features to extend the training context window size; and a bootstrapping framework that bridges our model with pluggable MLLMs without requiring annotation. We validate TGB across seven video benchmarks and demonstrate substantial performance improvements compared with prior MLLMs. Notably, our model, initially trained on sequences of four frames, effectively handles sequences up to 16 longer without sacrificing performance, highlighting its scalability and effectiveness in real-world applications. Our code is publicly available at https://github.com/bigai-nlco/VideoTGB, Comment: To appear at EMNLP 2024

Time series pre-training has recently garnered wide attention for its potential to reduce labeling expenses and benefit various downstream tasks. Prior methods are mainly based on pre-training techniques well-acknowledged in vision or language, such as masked modeling and contrastive learning. However, randomly masking time series or calculating series-wise similarity will distort or neglect inherent temporal correlations crucial in time series data. To emphasize temporal correlation modeling, this paper proposes TimeSiam as a simple but effective self-supervised pre-training framework for Time series based on Siamese networks. Concretely, TimeSiam pre-trains Siamese encoders to capture intrinsic temporal correlations between randomly sampled past and current subseries. With a simple data augmentation method (e.g.~masking), TimeSiam can benefit from diverse augmented subseries and learn internal time-dependent representations through a past-to-current reconstruction. Moreover, learnable lineage embeddings are also introduced to distinguish temporal distance between sampled series and further foster the learning of diverse temporal correlations. TimeSiam consistently outperforms extensive advanced pre-training baselines, demonstrating superior forecasting and classification capabilities across 13 standard benchmarks in both intra- and cross-domain scenarios.

We obtain the tail probability of generalized sub-Gaussian canonical processes. It can be viewed as a variant of the Bernstein-type inequality in the i.i.d case, and we further get a tighter bound of concentration inequality through uniformly randomized techniques. A concentration inequality for general functions involving independent random variables is also derived as an extension. As for applications, we derive convergence results for principal component analysis and the Rademacher complexities method., Comment: 25pages

We present a theory unifying the topological responses and anomalies of various gapless fermion systems exhibiting Fermi surfaces, including those with Berry phases, and nodal structures, which applies beyond non-interacting limit. As our key finding, we obtain a general approach to directly relate gapless fermions and topological insulators in phase space, including first- and higher-order insulators. Using this relation we show that the low-energy properties and response theories for gapless fermionic systems can be directly obtained without resorting to microscopic details. Our results provide a unified framework for describing such systems using well-developed theories from the study of topological phases of matter., Comment: 12 pages, 2 figures; references added

Recently we have witnessed the rapid development of video question answering models. However, most models can only handle simple videos in terms of temporal reasoning, and their performance tends to drop when answering temporal-reasoning questions on long and informative videos. To tackle this problem we propose STAIR, a Spatial-Temporal Reasoning model with Auditable Intermediate Results for video question answering. STAIR is a neural module network, which contains a program generator to decompose a given question into a hierarchical combination of several sub-tasks, and a set of lightweight neural modules to complete each of these sub-tasks. Though neural module networks are already widely studied on image-text tasks, applying them to videos is a non-trivial task, as reasoning on videos requires different abilities. In this paper, we define a set of basic video-text sub-tasks for video question answering and design a set of lightweight modules to complete them. Different from most prior works, modules of STAIR return intermediate outputs specific to their intentions instead of always returning attention maps, which makes it easier to interpret and collaborate with pre-trained models. We also introduce intermediate supervision to make these intermediate outputs more accurate. We conduct extensive experiments on several video question answering datasets under various settings to show STAIR's performance, explainability, compatibility with pre-trained models, and applicability when program annotations are not available. Code: https://github.com/yellow-binary-tree/STAIR, Comment: To appear in AAAI 2024

Higher-order topological insulators in two spatial dimensions display fractional corner charges. While fractional charges in one dimension are known to be captured by a many-body bulk invariant, computed by the Resta formula, a many-body bulk invariant for higher-order topology and the corresponding fractional corner charges remains elusive despite several attempts. Inspired by recent work by Tada and Oshikawa, we propose a well-defined many-body bulk invariant for $C_n$ symmetric higher-order topological insulators, which is valid for both non-interacting and interacting systems. Instead of relating them to the bulk quadrupole moment as was previously done, we show that in the presence of $C_n$ rotational symmetry, this bulk invariant can be directly identified with quantized fractional corner charges. In particular, we prove that the corner charge is quantized as $e/n$ with $C_n$ symmetry, leading to a $\mathbb{Z}_n$ classification for higher-order topological insulators in two dimensions., Comment: 12 pages, 7 figures, references updated

The Stable Diffusion Model (SDM) is a popular and efficient text-to-image (t2i) generation and image-to-image (i2i) generation model. Although there have been some attempts to reduce sampling steps, model distillation, and network quantization, these previous methods generally retain the original network architecture. Billion scale parameters and high computing requirements make the research of model architecture adjustment scarce. In this work, we first explore the computational redundancy part of the network, and then prune the redundancy blocks of the model and maintain the network performance through a progressive incubation strategy. Secondly, in order to maintaining the model performance, we add cross-layer multi-expert conditional convolution (CLME-Condconv) to the block pruning part to inherit the original convolution parameters. Thirdly, we propose a global-regional interactive (GRI) attention to speed up the computationally intensive attention part. Finally, we use semantic-aware supervision (SAS) to align the outputs of the teacher model and student model at the semantic level. Experiments show that this method can effectively train a lightweight model close to the performance of the original SD model, and effectively improve the model speed under limited resources. Experiments show that the proposed method can effectively train a light-weight model close to the performance of the original SD model, and effectively improve the model speed under limited resources. After acceleration, the UNet part of the model is 22% faster and the overall speed is 19% faster., Comment: Since the experimental part has not been added, we wish to withdraw the manuscript, and we hope to submit it after the experiment has been verified