Search

Generating natural and meaningful responses to communicate with multi-modal human inputs is a fundamental capability of Large Vision-Language Models(LVLMs). While current open-source LVLMs demonstrate promising performance in simplified scenarios such as single-turn single-image input, they fall short in real-world conversation scenarios such as following instructions in a long context history with multi-turn and multi-images. Existing LVLM benchmarks primarily focus on single-choice questions or short-form responses, which do not adequately assess the capabilities of LVLMs in real-world human-AI interaction applications. Therefore, we introduce MMDU, a comprehensive benchmark, and MMDU-45k, a large-scale instruction tuning dataset, designed to evaluate and improve LVLMs' abilities in multi-turn and multi-image conversations. We employ the clustering algorithm to ffnd the relevant images and textual descriptions from the open-source Wikipedia and construct the question-answer pairs by human annotators with the assistance of the GPT-4o model. MMDU has a maximum of 18k image+text tokens, 20 images, and 27 turns, which is at least 5x longer than previous benchmarks and poses challenges to current LVLMs. Our in-depth analysis of 15 representative LVLMs using MMDU reveals that open-source LVLMs lag behind closed-source counterparts due to limited conversational instruction tuning data. We demonstrate that ffne-tuning open-source LVLMs on MMDU-45k signiffcantly address this gap, generating longer and more accurate conversations, and improving scores on MMDU and existing benchmarks (MMStar: +1.1%, MathVista: +1.5%, ChartQA:+1.2%). Our contributions pave the way for bridging the gap between current LVLM models and real-world application demands. This project is available at https://github.com/Liuziyu77/MMDU., Comment: This project is available at https://github.com/Liuziyu77/MMDU

Recent semantic communication methods explore effective ways to expand the communication paradigm and improve the system performance of the communication systems. Nonetheless, the common problem of these methods is that the essence of semantics is not explicitly pointed out and directly utilized. A new epistemology suggests that synonymy, which is revealed as the fundamental feature of semantics, guides the establishment of the semantic information theory from a novel viewpoint. Building on this theoretical basis, this paper proposes a semantic arithmetic coding (SAC) method for semantic lossless compression using intuitive semantic synonymy. By constructing reasonable synonymous mappings and performing arithmetic coding procedures over synonymous sets, SAC can achieve higher compression efficiency for meaning-contained source sequences at the semantic level and thereby approximate the semantic entropy limits. Experimental results on edge texture map compression show an evident improvement in coding efficiency using SAC without semantic losses, compared to traditional arithmetic coding, which demonstrates its effectiveness., Comment: 6 pages, 4 figures. This paper is submitted to the 2024 IEEE International Symposium on Information Theory (ISIT 2024)

Semantic communication stands out as a highly promising avenue for future developments in communications. Theoretically, source compression coding based on semantics can achieve lower rates than Shannon entropy. This paper introduces a semantic Huffman coding built upon semantic information theory. By incorporating synonymous mapping and synonymous sets, semantic Huffman coding can achieve shorter average code lengths. Furthermore, we demonstrate that semantic Huffman coding theoretically have the capability to approximate semantic entropy. Experimental results indicate that, under the condition of semantic lossless, semantic Huffman coding exhibits clear advantages in compression efficiency over classical Huffman coding., Comment: 6 pages, 3 figures, this paper is submitted to the 2024 IEEE International Symposium on Information Theory (ISIT 2024)

In this paper, we introduce an algorithm for extracting topological data from translation invariant generalized Pauli stabilizer codes in two-dimensional systems, focusing on the analysis of anyon excitations and string operators. The algorithm applies to $\mathbb{Z}_d$ qudits, including instances where $d$ is a nonprime number. This capability allows the identification of topological orders that differ from the $\mathbb{Z}_d$ toric codes. It extends our understanding beyond the established theorem that Pauli stabilizer codes for $\mathbb{Z}_p$ qudits (with $p$ being a prime) are equivalent to finite copies of $\mathbb{Z}_p$ toric codes and trivial stabilizers. The algorithm is designed to determine all anyons and their string operators, enabling the computation of their fusion rules, topological spins, and braiding statistics. The method converts the identification of topological orders into computational tasks, including Gaussian elimination, the Hermite normal form, and the Smith normal form of truncated Laurent polynomials. Furthermore, the algorithm provides a systematic approach for studying quantum error-correcting codes. We apply it to various codes, such as self-dual CSS quantum codes modified from the 2d honeycomb color code and non-CSS quantum codes that contain the double semion topological order or the six-semion topological order., Comment: main text+appendix 29 pages, 27 figures

In-band full-duplex (IBFD) is a theoretically effective solution to increase the overall throughput for the future wireless communications system by enabling transmission and reception over the same time-frequency resources. However, reliable source reconstruction remains a great challenge in the practical IBFD systems due to the non-ideal elimination of the self-interference and the inherent limitations of the separate source and channel coding methods. On the other hand, artificial intelligence-enabled semantic communication can provide a viable direction for the optimization of the IBFD system. This article introduces a novel IBFD paradigm with the guidance of semantic communication called semantics-division duplexing (SDD). It utilizes semantic domain processing to further suppress self-interference, distinguish the expected semantic information, and recover the desired sources. Further integration of the digital and semantic domain processing can be implemented so as to achieve intelligent and concise communications. We present the advantages of the SDD paradigm with theoretical explanations and provide some visualized results to verify its effectiveness., Comment: 9 pages, 5 figures, Accepted by IEEE Wireless Communications Magazine

In this paper, we design a new class of high-efficiency deep joint source-channel coding methods to achieve end-to-end video transmission over wireless channels. The proposed methods exploit nonlinear transform and conditional coding architecture to adaptively extract semantic features across video frames, and transmit semantic feature domain representations over wireless channels via deep joint source-channel coding. Our framework is collected under the name deep video semantic transmission (DVST). In particular, benefiting from the strong temporal prior provided by the feature domain context, the learned nonlinear transform function becomes temporally adaptive, resulting in a richer and more accurate entropy model guiding the transmission of current frame. Accordingly, a novel rate adaptive transmission mechanism is developed to customize deep joint source-channel coding for video sources. It learns to allocate the limited channel bandwidth within and among video frames to maximize the overall transmission performance. The whole DVST design is formulated as an optimization problem whose goal is to minimize the end-to-end transmission rate-distortion performance under perceptual quality metrics or machine vision task performance metrics. Across standard video source test sequences and various communication scenarios, experiments show that our DVST can generally surpass traditional wireless video coded transmission schemes. The proposed DVST framework can well support future semantic communications due to its video content-aware and machine vision task integration abilities., Comment: published in IEEE JSAC