Search

Advanced Multimodal Large Language Models (MLLMs) struggle with recent Knowledge-based VQA tasks, such as INFOSEEK and Encyclopedic-VQA, due to their limited and frozen knowledge scope, often leading to ambiguous and inaccurate responses. Thus, multimodal Retrieval-Augmented Generation (mRAG) is naturally introduced to provide MLLMs with comprehensive and up-to-date knowledge, effectively expanding the knowledge scope. However, current mRAG methods have inherent drawbacks, including: 1) Performing retrieval even when external knowledge is not needed. 2) Lacking of identification of evidence that supports the query. 3) Increasing model complexity due to additional information filtering modules or rules. To address these shortcomings, we propose a novel generalized framework called \textbf{m}ultimodal \textbf{R}etrieval-\textbf{R}eflection-\textbf{A}ugmented \textbf{G}eneration (mR$^2$AG), which achieves adaptive retrieval and useful information localization to enable answers through two easy-to-implement reflection operations, preventing high model complexity. In mR$^2$AG, Retrieval-Reflection is designed to distinguish different user queries and avoids redundant retrieval calls, and Relevance-Reflection is introduced to guide the MLLM in locating beneficial evidence of the retrieved content and generating answers accordingly. In addition, mR$^2$AG can be integrated into any well-trained MLLM with efficient fine-tuning on the proposed mR$^2$AG Instruction-Tuning dataset (mR$^2$AG-IT). mR$^2$AG significantly outperforms state-of-the-art MLLMs (e.g., GPT-4v/o) and RAG-based MLLMs on INFOSEEK and Encyclopedic-VQA, while maintaining the exceptional capabilities of base MLLMs across a wide range of Visual-dependent tasks.

Recent studies have highlighted the significant potential of Large Language Models (LLMs) as zero-shot relevance rankers. These methods predominantly utilize prompt learning to assess the relevance between queries and documents by generating a ranked list of potential documents. Despite their promise, the substantial costs associated with LLMs pose a significant challenge for their direct implementation in commercial search systems. To overcome this barrier and fully exploit the capabilities of LLMs for text ranking, we explore techniques to transfer the ranking expertise of LLMs to a more compact model similar to BERT, using a ranking loss to enable the deployment of less resource-intensive models. Specifically, we enhance the training of LLMs through Continued Pre-Training, taking the query as input and the clicked title and summary as output. We then proceed with supervised fine-tuning of the LLM using a rank loss, assigning the final token as a representative of the entire sentence. Given the inherent characteristics of autoregressive language models, only the final token can encapsulate all preceding tokens. Additionally, we introduce a hybrid point-wise and margin MSE loss to transfer the ranking knowledge from LLMs to smaller models like BERT. This method creates a viable solution for environments with strict resource constraints. Both offline and online evaluations have confirmed the efficacy of our approach, and our model has been successfully integrated into a commercial web search engine as of February 2024., Comment: Arxiv Version

Recently, Linear Complementary Dual (LCD) codes have garnered substantial interest within coding theory research due to their diverse applications and favorable attributes. This paper directs its attention to the construction of binary and ternary LCD codes leveraging curiosity-driven reinforcement learning (RL). By establishing reward and devising well-reasoned mappings from actions to states, it aims to facilitate the successful synthesis of binary or ternary LCD codes. Experimental results indicate that LCD codes constructed using RL exhibit slightly superior error-correction performance compared to those conventionally constructed LCD codes and those developed via standard RL methodologies. The paper introduces novel binary and ternary LCD codes with enhanced minimum distance bounds. Finally, it showcases how Random Network Distillation aids agents in exploring beyond local optima, enhancing the overall performance of the models without compromising convergence., Comment: 17 pages, just accepted by JSSC

International trading networks significantly influence global economic conditions and environmental outcomes. A notable imbalance between economic gains and emissions transfers persists, manifesting as carbon inequality. This study introduces a novel metric, the Ecological Economic Equality Index, integrated with complex network dynamics analysis, to quantitatively evaluate the evolving roles within the global trading network and to pinpoint inequities in trade relationships from 1995 to 2022. Utilising high spatiotemporal resolution data from the Environmentally Extended Multi-regional Input-output model, our findings reveal a widening disparity in carbon inequality and dynamic patterns. This analysis emphasises the gap in regional carbon inequality and identifies unequal trade. The study underscores that carbon inequality is a critical challenge affecting both developing and developed regions, demanding widespread attention and action.

In this paper we investigate the issues regarding the convergence of the Policy Iteration Algorithm(PIA) for a class of general continuous-time entropy-regularized stochastic control problems. In particular, instead of employing sophisticated PDE estimates for the iterative PDEs involved in the PIA (see, e.g., Huang-Wang-Zhou(2023)), we shall provide a simple proof from scratch for the convergence of the PIA. Our approach builds on probabilistic representation formulae for solutions of PDEs and their derivatives. Moreover, in the infinite horizon model with large discount factor and in the finite horizon model, the similar arguments lead to the exponential rate of convergence of PIA without tear. Finally, with some extra efforts we show that our approach can also be extended to the case when diffusion contains control, in the one dimensional setting but without much extra constraints on the coefficients. We believe that these results are new in the literature., Comment: In this version, we have added results on convergence and rate of convergence for the diffusion control problem in the scalar case

The Environmental Extended Multi-Regional Input-Output analysis is the predominant framework in Ecological Economics for assessing the environmental impact of economic activities. This paper introduces ExioML, the first Machine Learning benchmark dataset designed for sustainability analysis, aimed at lowering barriers and fostering collaboration between Machine Learning and Ecological Economics research. A crucial greenhouse gas emission regression task was conducted to evaluate sectoral sustainability and demonstrate the usability of the dataset. We compared the performance of traditional shallow models with deep learning models, utilizing a diverse Factor Accounting table and incorporating various categorical and numerical features. Our findings reveal that ExioML, with its high usability, enables deep and ensemble models to achieve low mean square errors, establishing a baseline for future Machine Learning research. Through ExioML, we aim to build a foundational dataset supporting various Machine Learning applications and promote climate actions and sustainable investment decisions.

In this paper, we mainly focus on the set-valued (stochastic) analysis on the space of convex, closed, but possibly unbounded sets, and try to establish a useful theoretical framework for studying the set-valued stochastic differential equations with unbounded coefficients. The space that we will be focusing on are convex, closed sets that are "generated" by a given cone, in the sense that the Hausdorff distance of all elements to the "generating" cone is finite. Such space should in particular include the so-called "upper sets", and has many useful cases in finance, such as the well-known set-valued risk measures, as well as the solvency cone in some super-hedging problems. We shall argue that, for such a special class of unbounded sets, under some conditions, the cancellation law is still valid, eliminating a major obstacle for extending the set-valued analysis to non-compact sets. We shall establish some basic algebraic and topological properties of such spaces, and show that some standard techniques will again be valid in studying the set-valued SDEs with unbounded (drift) coefficients which, to the best of our knowledge, is new.

Non-Intrusive Load Monitoring (NILM) is pivotal in today's energy landscape, offering vital solutions for energy conservation and efficient management. Its growing importance in enhancing energy savings and understanding consumer behavior makes it a pivotal technology for addressing global energy challenges. This paper delivers an in-depth review of NILM, highlighting its critical role in smart homes and smart grids. The significant contributions of this study are threefold: Firstly, it compiles a comprehensive global dataset table, providing a valuable tool for researchers and engineers to select appropriate datasets for their NILM studies. Secondly, it categorizes NILM approaches, simplifying the understanding of various algorithms by focusing on technologies, label data requirements, feature usage, and monitoring states. Lastly, by identifying gaps in current NILM research, this work sets a clear direction for future studies, discussing potential areas of innovation., Comment: a comprehensive summary with a dataset list

Enhancing the attribution in large language models (LLMs) is a crucial task. One feasible approach is to enable LLMs to cite external sources that support their generations. However, existing datasets and evaluation methods in this domain still exhibit notable limitations. In this work, we formulate the task of attributed query-focused summarization (AQFS) and present WebCiteS, a Chinese dataset featuring 7k human-annotated summaries with citations. WebCiteS derives from real-world user queries and web search results, offering a valuable resource for model training and evaluation. Prior works in attribution evaluation do not differentiate between groundedness errors and citation errors. They also fall short in automatically verifying sentences that draw partial support from multiple sources. We tackle these issues by developing detailed metrics and enabling the automatic evaluator to decompose the sentences into sub-claims for fine-grained verification. Our comprehensive evaluation of both open-source and proprietary models on WebCiteS highlights the challenge LLMs face in correctly citing sources, underscoring the necessity for further improvement. The dataset and code will be open-sourced to facilitate further research in this crucial field., Comment: 20 pages, 7 figures, accepted to ACL 2024 main conference

In this paper, we reanalyze the top-quark pair production at the next-to-next-to-leading order (NNLO) in QCD at future $e^+e^-$ colliders by using the Principle of Maximum Conformality (PMC) method. The PMC renormalization scales in $\alpha_s$ are determined by absorbing the non-conformal $\beta$ terms by recursively using the Renormalization Group Equation (RGE). Unlike the conventional scale-setting method of fixing the scale at the center-of-mass energy $\mu_r=\sqrt{s}$, the determined PMC scale $Q_\star$ is far smaller than the $\sqrt{s}$ and increases with the $\sqrt{s}$, yielding the correct physical behavior for the top-quark pair production process. Moreover, the convergence of the pQCD series for the top-quark pair production is greatly improved due to the elimination of the renormalon divergence. For a typical collision energy of $\sqrt{s}=500$ GeV, the PMC scale is $Q_\star=107$ GeV; the QCD correction factor $K$ for conventional results is $K\sim1+0.1244^{+0.0102+0.0012}_{-0.0087-0.0011}+0.0184^{-0.0086+0.0002}_{+0.0061-0.0003}$, where the first error is caused by varying the scale $\mu_r\in[\sqrt{s}/2, 2\sqrt{s}]$ and the second error is from the top-quark mass $\Delta{m_t}=\pm0.7$ GeV. After applying the PMC, the renormalization scale uncertainty is eliminated and the QCD correction factor $K$ is improved to $K\sim 1+0.1507^{+0.0015}_{-0.0015}-0.0057^{+0.0001}_{-0.0000}$, where the error is from the top-quark mass $\Delta{m_t}=\pm0.7$ GeV. The PMC improved predictions for the top-quark pair production are helpful for detailed studies of properties of the top-quark at future $e^+e^-$ colliders., Comment: 8 pages, 3 figures. Accepted by Chinese Physics C

Vision-Language Models pre-trained on large-scale image-text datasets have shown superior performance in downstream tasks such as image retrieval. Most of the images for pre-training are presented in the form of open domain common-sense visual elements. Differently, video covers in short video search scenarios are presented as user-originated contents that provide important visual summaries of videos. In addition, a portion of the video covers come with manually designed cover texts that provide semantic complements. In order to fill in the gaps in short video cover data, we establish the first large-scale cover-text benchmark for Chinese short video search scenarios. Specifically, we release two large-scale datasets CBVS-5M/10M to provide short video covers, and the manual fine-labeling dataset CBVS-20K to provide real user queries, which serves as an image-text benchmark test in the Chinese short video search field. To integrate the semantics of cover text in the case of modality missing, we propose UniCLIP where cover texts play a guiding role during training, however are not relied upon by inference. Extensive evaluation on CBVS-20K demonstrates the excellent performance of our proposal. UniCLIP has been deployed to Tencent's online video search systems with hundreds of millions of visits and achieved significant gains. The dataset and code are available at https://github.com/QQBrowserVideoSearch/CBVS-UniCLIP.

In this paper, a viscous shock wave under space-periodic perturbation of 1-D isentropic Navier-Stokes system in the half space is investigated. It is shown that if the initial periodic perturbation around the viscous shock wave is small, then the solution time asymptotically tends to a viscous shock wave with a shift partially determined by the periodic oscillations. Moreover, the strength of {the} shock wave could be arbitrarily large. This result essentially improves the previous work " A. Matsumura, M. Mei, Convergence to travelling fronts of solutions of the p-system with viscosity in the presence of a boundary. Arch. Ration. Mech. Anal. 146 (1999), no. 1, 1-22." where the strength of shock wave is sufficiently small and the initial periodic oscillations vanish., Comment: 2 figures

Information retrieval in real-time search presents unique challenges distinct from those encountered in classical web search. These challenges are particularly pronounced due to the rapid change of user search intent, which is influenced by the occurrence and evolution of breaking news events, such as earthquakes, elections, and wars. Previous dense retrieval methods, which primarily focused on static semantic representation, lack the capacity to capture immediate search intent, leading to inferior performance in retrieving the most recent event-related documents in time-sensitive scenarios. To address this issue, this paper expands the query with event information that represents real-time search intent. The Event information is then integrated with the query through a cross-attention mechanism, resulting in a time-context query representation. We further enhance the model's capacity for event representation through multi-task training. Since publicly available datasets such as MS-MARCO do not contain any event information on the query side and have few time-sensitive queries, we design an automatic data collection and annotation pipeline to address this issue, which includes ModelZoo-based Coarse Annotation and LLM-driven Fine Annotation processes. In addition, we share the training tricks such as two-stage training and hard negative sampling. Finally, we conduct a set of offline experiments on a million-scale production dataset to evaluate our approach and deploy an A/B testing in a real online system to verify the performance. Extensive experimental results demonstrate that our proposed approach significantly outperforms existing state-of-the-art baseline methods.

In this paper, we study the optimal dividend problem under the continuous time diffusion model with the dividend rate being restricted in a given finite interval. Unlike the standard literature, we shall particularly be interested in the case when the parameters (e.g. drift and diffusion coefficients) of the model are not specified so that the optimal control cannot be explicitly determined. We therefore follow the recently developed method via the Reinforcement Learning (RL) to find the optimal strategy. Specifically, we shall design a corresponding RL-type entropy-regularized exploratory control problem, which randomize the control actions, and balance the exploitation and exploration. We shall first carry out a theoretical analysis of the new relaxed control problem and prove that the value function is the unique bounded classical solution to the corresponding HJB equation. We will then use a policy improvement argument, along with policy evaluation devices (e.g., Temporal Difference (TD)-based algorithm and Martingale Loss (ML)-algorithms) to construct approximating sequences of the optimal strategy. We present some numerical results using different parametrization families for the cost functional to illustrate the effectiveness of the approximation schemes.

Current Scene text image super-resolution approaches primarily focus on extracting robust features, acquiring text information, and complex training strategies to generate super-resolution images. However, the upsampling module, which is crucial in the process of converting low-resolution images to high-resolution ones, has received little attention in existing works. To address this issue, we propose the Pixel Adapter Module (PAM) based on graph attention to address pixel distortion caused by upsampling. The PAM effectively captures local structural information by allowing each pixel to interact with its neighbors and update features. Unlike previous graph attention mechanisms, our approach achieves 2-3 orders of magnitude improvement in efficiency and memory utilization by eliminating the dependency on sparse adjacency matrices and introducing a sliding window approach for efficient parallel computation. Additionally, we introduce the MLP-based Sequential Residual Block (MSRB) for robust feature extraction from text images, and a Local Contour Awareness loss ($\mathcal{L}_{lca}$) to enhance the model's perception of details. Comprehensive experiments on TextZoom demonstrate that our proposed method generates high-quality super-resolution images, surpassing existing methods in recognition accuracy. For single-stage and multi-stage strategies, we achieved improvements of 0.7\% and 2.6\%, respectively, increasing the performance from 52.6\% and 53.7\% to 53.3\% and 56.3\%. The code is available at https://github.com/wenyu1009/RTSRN.