Search

The recent surge of Multimodal Large Language Models (MLLMs) has fundamentally reshaped the landscape of AI research and industry, shedding light on a promising path toward the next AI milestone. However, significant challenges remain preventing MLLMs from being practical in real-world applications. The most notable challenge comes from the huge cost of running an MLLM with a massive number of parameters and extensive computation. As a result, most MLLMs need to be deployed on high-performing cloud servers, which greatly limits their application scopes such as mobile, offline, energy-sensitive, and privacy-protective scenarios. In this work, we present MiniCPM-V, a series of efficient MLLMs deployable on end-side devices. By integrating the latest MLLM techniques in architecture, pretraining and alignment, the latest MiniCPM-Llama3-V 2.5 has several notable features: (1) Strong performance, outperforming GPT-4V-1106, Gemini Pro and Claude 3 on OpenCompass, a comprehensive evaluation over 11 popular benchmarks, (2) strong OCR capability and 1.8M pixel high-resolution image perception at any aspect ratio, (3) trustworthy behavior with low hallucination rates, (4) multilingual support for 30+ languages, and (5) efficient deployment on mobile phones. More importantly, MiniCPM-V can be viewed as a representative example of a promising trend: The model sizes for achieving usable (e.g., GPT-4V) level performance are rapidly decreasing, along with the fast growth of end-side computation capacity. This jointly shows that GPT-4V level MLLMs deployed on end devices are becoming increasingly possible, unlocking a wider spectrum of real-world AI applications in the near future., Comment: preprint

For noisy environments, ensuring the robustness of keyword spotting (KWS) systems is essential. While much research has focused on noisy KWS, less attention has been paid to multi-talker mixed speech scenarios. Unlike the usual cocktail party problem where multi-talker speech is separated using speaker clues, the key challenge here is to extract the target speech for KWS based on text clues. To address it, this paper proposes a novel Text-aware Permutation Determinization Training method for multi-talker KWS with a clue-based Speech Separation front-end (TPDT-SS). Our research highlights the critical role of SS front-ends and shows that incorporating keyword-specific clues into these models can greatly enhance the effectiveness. TPDT-SS shows remarkable success in addressing permutation problems in mixed keyword speech, thereby greatly boosting the performance of the backend. Additionally, fine-tuning our system on unseen mixed speech results in further performance improvement., Comment: Accepted by INTERSPEECH2024

This paper addresses the following problem. \begin{equation} \left\{ \begin{array}{lr} -{\Delta}u=\lambda I_\alpha*_\Omega u+|u|^{2^*-2}u\mbox{ in }\Omega ,\nonumber u\in H_0^1(\Omega).\nonumber \end{array} \right. \end{equation} Here, $\Omega$ is a bounded domain in $\mathbb{R}^N$ with $N\geq3$, $2^*=\frac{2N}{N-2}$, $\lambda\in\mathbb{R}$, $\lambda\in(0,N)$, $I_\alpha$ is the Riesz potential and \begin{align} I_\alpha*_\Omega u(x):=\int_\Omega \frac{\Gamma(\frac{N-\alpha}{2})}{\Gamma(\frac{\alpha}{2})\pi^\frac{N}{2}2^\alpha|x-y|^{N-\alpha}} u(y)dy. \nonumber \end{align} We study the non-existence, existence and multiplicity results. Our argument combines Brezis-Nirenberg's method with the regularity results involving potential terms. Especially, we study the following nonlocal eigenvalue problem. \begin{equation} \left\{ \begin{array}{lr} -{\Delta}u=\lambda I_\alpha*_\Omega u\mbox{ in }\Omega ,\nonumber \lambda\in\mathbb{R},\,u\in H_0^1(\Omega).\nonumber \end{array} \right. \end{equation}, Comment: 25 pages

Word embedding, a high-dimensional (HD) numerical representation of words generated by machine learning models, has been used for different natural language processing tasks, e.g., translation between two languages. Recently, there has been an increasing trend of transforming the HD embeddings into a latent space (e.g., via autoencoders) for further tasks, exploiting various merits the latent representations could bring. To preserve the embeddings' quality, these works often map the embeddings into an even higher-dimensional latent space, making the already complicated embeddings even less interpretable and consuming more storage space. In this work, we borrow the idea of $\beta$VAE to regularize the HD latent space. Our regularization implicitly condenses information from the HD latent space into a much lower-dimensional space, thus compressing the embeddings. We also show that each dimension of our regularized latent space is more semantically salient, and validate our assertion by interactively probing the encoding-level of user-proposed semantics in the dimensions. To the end, we design a visual analytics system to monitor the regularization process, explore the HD latent space, and interpret latent dimensions' semantics. We validate the effectiveness of our embedding regularization and interpretation approach through both quantitative and qualitative evaluations.

Designing an efficient keyword spotting (KWS) system that delivers exceptional performance on resource-constrained edge devices has long been a subject of significant attention. Existing KWS search algorithms typically follow a frame-synchronous approach, where search decisions are made repeatedly at each frame despite the fact that most frames are keyword-irrelevant. In this paper, we propose TDT-KWS, which leverages token-and-duration Transducers (TDT) for KWS tasks. We also propose a novel KWS task-specific decoding algorithm for Transducer-based models, which supports highly effective frame-asynchronous keyword search in streaming speech scenarios. With evaluations conducted on both the public Hey Snips and self-constructed LibriKWS-20 datasets, our proposed KWS-decoding algorithm produces more accurate results than conventional ASR decoding algorithms. Additionally, TDT-KWS achieves on-par or better wake word detection performance than both RNN-T and traditional TDT-ASR systems while achieving significant inference speed-up. Furthermore, experiments show that TDT-KWS is more robust to noisy environments compared to RNN-T KWS., Comment: Accepted by ICASSP2024

Implicit Neural representations (INRs) are widely used for scientific data reduction and visualization by modeling the function that maps a spatial location to a data value. Without any prior knowledge about the spatial distribution of values, we are forced to sample densely from INRs to perform visualization tasks like iso-surface extraction which can be very computationally expensive. Recently, range analysis has shown promising results in improving the efficiency of geometric queries, such as ray casting and hierarchical mesh extraction, on INRs for 3D geometries by using arithmetic rules to bound the output range of the network within a spatial region. However, the analysis bounds are often too conservative for complex scientific data. In this paper, we present an improved technique for range analysis by revisiting the arithmetic rules and analyzing the probability distribution of the network output within a spatial region. We model this distribution efficiently as a Gaussian distribution by applying the central limit theorem. Excluding low probability values, we are able to tighten the output bounds, resulting in a more accurate estimation of the value range, and hence more accurate identification of iso-surface cells and more efficient iso-surface extraction on INRs. Our approach demonstrates superior performance in terms of the iso-surface extraction time on four datasets compared to the original range analysis method and can also be generalized to other geometric query tasks., Comment: Accepted to IEEE Transactions on Visualization and Computer Graphics, presented in VIS 2024

Reconstructing visual stimuli from functional Magnetic Resonance Imaging (fMRI) based on Latent Diffusion Models (LDM) provides a fine-grained retrieval of the brain. A challenge persists in reconstructing a cohesive alignment of details (such as structure, background, texture, color, etc.). Moreover, LDMs would generate different image results even under the same conditions. For these, we first uncover the neuroscientific perspective of LDM-based methods that is top-down creation based on pre-trained knowledge from massive images but lack of detail-driven bottom-up perception resulting in unfaithful details. We propose NeuralDiffuser which introduces primary visual feature guidance to provide detail cues in the form of gradients, extending the bottom-up process for LDM-based methods to achieve faithful semantics and details. We also developed a novel guidance strategy to ensure the consistency of repeated reconstructions rather than a variety of results. We obtain the state-of-the-art performance of NeuralDiffuser on the Natural Senses Dataset (NSD), which offers more faithful details and consistent results., Comment: The implementation error lead to incorrect results in experiment

As deep neural networks (DNNs) grow in complexity and size, the resultant increase in communication overhead during distributed training has become a significant bottleneck, challenging the scalability of distributed training systems. Existing solutions, while aiming to mitigate this bottleneck through worker-level compression and in-network aggregation, fall short due to their inability to efficiently reconcile the trade-offs between compression effectiveness and computational overhead, hindering overall performance and scalability. In this paper, we introduce a novel compression algorithm that effectively merges worker-level compression with in-network aggregation. Our solution is both homomorphic, allowing for efficient in-network aggregation without CPU/GPU processing, and lossless, ensuring no compromise on training accuracy. Theoretically optimal in compression and computational efficiency, our approach is empirically validated across diverse DNN models such as NCF, LSTM, VGG19, and BERT-base, showing up to a 6.33$\times$ improvement in aggregation throughput and a 3.74$\times$ increase in per-iteration training speed.

Persuasion, as one of the crucial abilities in human communication, has garnered extensive attention from researchers within the field of intelligent dialogue systems. We humans tend to persuade others to change their viewpoints, attitudes or behaviors through conversations in various scenarios (e.g., persuasion for social good, arguing in online platforms). Developing dialogue agents that can persuade others to accept certain standpoints is essential to achieving truly intelligent and anthropomorphic dialogue system. Benefiting from the substantial progress of Large Language Models (LLMs), dialogue agents have acquired an exceptional capability in context understanding and response generation. However, as a typical and complicated cognitive psychological system, persuasive dialogue agents also require knowledge from the domain of cognitive psychology to attain a level of human-like persuasion. Consequently, the cognitive strategy-enhanced persuasive dialogue agent (defined as CogAgent), which incorporates cognitive strategies to achieve persuasive targets through conversation, has become a predominant research paradigm. To depict the research trends of CogAgent, in this paper, we first present several fundamental cognitive psychology theories and give the formalized definition of three typical cognitive strategies, including the persuasion strategy, the topic path planning strategy, and the argument structure prediction strategy. Then we propose a new system architecture by incorporating the formalized definition to lay the foundation of CogAgent. Representative works are detailed and investigated according to the combined cognitive strategy, followed by the summary of authoritative benchmarks and evaluation metrics. Finally, we summarize our insights on open issues and future directions of CogAgent for upcoming researchers., Comment: 36 pages, 6 figures

In this paper, we study the sign-changing radial solutions of the following coupled Schr\"odinger system \begin{equation} \left\{ \begin{array}{lr} -{\Delta}u_j+\lambda_j u_j=\mu_j u_j^3+\sum_{i\neq j}\beta_{ij} u_i^2 u_j \,\,\,\,\,\,\,\, \mbox{in }B_1 ,\nonumber u_j\in H_{0,r}^1(B_1)\mbox{ for }j=1,\cdots,N.\nonumber \end{array} \right. \end{equation} Here, $\lambda_j,\,\mu_j>0$ and $\beta_{ij}=\beta_{ji}$ are constants for $i,j=1,\cdots,N$ and $i\neq j$. $B_1$ denotes the unit ball in the Euclidean space $\mathbb{R}^3$ centred at the origin. For any $P_1,\cdots,P_N\in\mathbb{N}$, we prove the uniqueness of the radial solution $(u_1,\cdots,u_j)$ with $u_j$ changes its sign exactly $P_j$ times for any $j=1,\cdots,N$ in the following case: $\lambda_j\geq1$ and $|\beta_{ij}|$ are small for $i,j=1,\cdots,N$ and $i\neq j$. New Liouville-type theorems and boundedness results are established for this purpose., Comment: 13 pages

Data attribution methods trace model behavior back to its training dataset, offering an effective approach to better understand ''black-box'' neural networks. While prior research has established quantifiable links between model output and training data in diverse settings, interpreting diffusion model outputs in relation to training samples remains underexplored. In particular, diffusion models operate over a sequence of timesteps instead of instantaneous input-output relationships in previous contexts, posing a significant challenge to extend existing frameworks to diffusion models directly. Notably, we present Diffusion-TracIn that incorporates this temporal dynamics and observe that samples' loss gradient norms are highly dependent on timestep. This trend leads to a prominent bias in influence estimation, and is particularly noticeable for samples trained on large-norm-inducing timesteps, causing them to be generally influential. To mitigate this effect, we introduce Diffusion-ReTrac as a re-normalized adaptation that enables the retrieval of training samples more targeted to the test sample of interest, facilitating a localized measurement of influence and considerably more intuitive visualization. We demonstrate the efficacy of our approach through various evaluation metrics and auxiliary tasks, reducing the amount of generally influential samples to $\frac{1}{3}$ of its original quantity., Comment: Accepted at TMLR. Code available at https://github.com/txie1/diffusion-ReTrac

Metallic Glasses (MGs) are widely used materials that are stronger than steel while being shapeable as plastic. While understanding the structure-property relationship of MGs remains a challenge in materials science, studying their energy barriers (EBs) as an intermediary step shows promise. In this work, we utilize Graph Neural Networks (GNNs) to model MGs and study EBs. We contribute a new dataset for EB prediction and a novel Symmetrized GNN (SymGNN) model that is E(3)-invariant in expectation. SymGNN handles invariance by aggregating over orthogonal transformations of the graph structure. When applied to EB prediction, SymGNN are more accurate than molecular dynamics (MD) local-sampling methods and other machine-learning models. Compared to precise MD simulations, SymGNN reduces the inference time on new MGs from roughly 41 days to less than one second. We apply explanation algorithms to reveal the relationship between structures and EBs. The structures that we identify through explanations match the medium-range order (MRO) hypothesis and possess unique topological properties. Our work enables effective prediction and interpretation of MG EBs, bolstering material science research., Comment: ICML 2024. Code available at https://github.com/haoyuli02/SymGNN

The paper deals with the existence of non-radial solutions for an $N$-coupled nonlinear elliptic system. In the repulsive regime with some structure conditions on the coupling and for each symmetric subspace of rotation symmetry, we prove the existence of an infinite sequence of non-radial positive solutions and an infinite sequence of non-radial nodal solutions.

Membership (membership query / membership testing) is a fundamental problem across databases, networks and security. However, previous research has primarily focused on either approximate solutions, such as Bloom Filters, or exact methods, like perfect hashing and dictionaries, without attempting to develop a an integral theory. In this paper, we propose a unified and complete theory, namely chain rule, for general membership problems, which encompasses both approximate and exact membership as extreme cases. Building upon the chain rule, we introduce a straightforward yet versatile algorithm framework, namely ChainedFilter, to combine different elementary filters without losing information. Our evaluation results demonstrate that ChainedFilter performs well in many applications: (1) it requires only 26% additional space over the theoretical lower bound for implicit static dictionary, (2) it requires only 0.22 additional bit per item over the theoretical lower bound for lossless data compression, (3) it reduces up to 31% external memory access than raw Cuckoo Hashing, (4) it reduces up to 36% P99 tail point query latency than Bloom Filter under the same space cost in RocksDB database, and (5) it reduces up to 99.1% filter space than original Learned Bloom Filter., Comment: SIGMOD 2024

Channel Reciprocity-based Key Generation (CRKG) exploits reciprocal channel randomness to establish shared secret keys between wireless terminals. This new security technique is expected to complement existing cryptographic techniques for secret key distribution of future wireless networks. In this paper, we present a new attack, reconfigurable intelligent surface (RIS) jamming, and show that an attacker can prevent legitimate users from agreeing on the same key by deploying a malicious RIS to break channel reciprocity. Specifically, we elaborate on three examples to implement the RIS jamming attack: Using active nonreciprocal circuits, performing time-varying controls, and reducing the signal-to-noise ratio. The attack effect is then studied by formulating the secret key rate with a relationship to the deployment of RIS. To resist such RIS jamming attacks, we propose a countermeasure that exploits wideband signals for multipath separation. The malicious RIS path is distinguished from all separated channel paths, and thus the countermeasure is referred to as contaminated path removal-based CRKG(CRP-CRKG). We present simulation results, showing that legitimate users under RIS jamming are still able to generate secret keys from the remaining paths. We also experimentally demonstrate the RIS jamming attack by using commodity Wi-Fi devices in conjunction with a fabricated RIS prototype. In our experiments, we were able to increase the average bit disagreement ratio (BDR) of raw secret keys by 20%. Further, we successfully demonstrate the proposed CRP-CRKG countermeasure to tackle RIS jamming in wideband systems as long as the source of randomness and the RIS propagation paths are separable., Comment: 15 pages, 14 figures

The anti-interference capability of wireless links is a physical layer problem for edge computing. Although convolutional codes have inherent error correction potential due to the redundancy introduced in the data, the performance of the convolutional code is drastically degraded due to multipath effects on the channel. In this paper, we propose the use of a Hidden Markov Model (HMM) for the reconstruction of convolutional codes and decoding by the Viterbi algorithm. Furthermore, to implement soft-decision decoding, the observation of HMM is replaced by Gaussian mixture models (GMM). Our method provides superior error correction potential than the standard method because the model parameters contain channel state information (CSI). We evaluated the performance of the method compared to standard Viterbi decoding by numerical simulation. In the multipath channel, the hybrid HMM decoder can achieve a performance gain of 4.7 dB and 2 dB when using hard-decision and soft-decision decoding, respectively. The HMM decoder also achieves significant performance gains for the RSC code, suggesting that the method could be extended to turbo codes., Comment: 12 pages, 8 figures