Multimodal reasoning is a challenging task that requires models to reason across multiple modalities to answer questions. Existing approaches have made progress by incorporating language and visual modalities into a two-stage reasoning framework, separating rationale generation from answer inference. However, these approaches often fall short due to the inadequate quality of the generated rationales. In this work, we delve into the importance of rationales in model reasoning. We observe that when rationales are completely accurate, the model's accuracy significantly improves, highlighting the need for high-quality rationale generation. Motivated by this, we propose MC-CoT, a self-consistency training strategy that generates multiple rationales and answers, subsequently selecting the most accurate through a voting process. This approach not only enhances the quality of generated rationales but also leads to more accurate and robust answers. Through extensive experiments, we demonstrate that our approach significantly improves model performance across various benchmarks. Remarkably, we show that even smaller base models, when equipped with our proposed approach, can achieve results comparable to those of larger models, illustrating the potential of our approach in harnessing the power of rationales for improved multimodal reasoning. The code is available at //github.com/chengtan9907/mc-cot.
相關內容
ACM/IEEE第23屆模型驅動工程語言和系統國際會議，是模型驅動軟件和系統工程的首要會議系列，由ACM-SIGSOFT和IEEE-TCSE支持組織。自1998年以來，模型涵蓋了建模的各個方面，從語言和方法到工具和應用程序。模特的參加者來自不同的背景，包括研究人員、學者、工程師和工業專業人士。MODELS 2019是一個論壇，參與者可以圍繞建模和模型驅動的軟件和系統交流前沿研究成果和創新實踐經驗。今年的版本將為建模社區提供進一步推進建模基礎的機會，并在網絡物理系統、嵌入式系統、社會技術系統、云計算、大數據、機器學習、安全、開源等新興領域提出建模的創新應用以及可持續性。
官網鏈接： ·
OCR
·
MoDELS
·
視覺問答
·
Gemini
·
Large models have recently played a dominant role in natural language processing and multimodal vision-language learning. However, their effectiveness in text-related visual tasks remains relatively unexplored. In this paper, we conducted a comprehensive evaluation of Large Multimodal Models, such as GPT4V and Gemini, in various text-related visual tasks including Text Recognition, Scene Text-Centric Visual Question Answering (VQA), Document-Oriented VQA, Key Information Extraction (KIE), and Handwritten Mathematical Expression Recognition (HMER). To facilitate the assessment of Optical Character Recognition (OCR) capabilities in Large Multimodal Models, we propose OCRBench, a comprehensive evaluation benchmark.Our study encompasses 29 datasets, making it the most comprehensive OCR evaluation benchmark available. Furthermore, our study reveals both the strengths and weaknesses of these models, particularly in handling multilingual text, handwritten text, non-semantic text, and mathematical expression recognition. Most importantly, the baseline results showcased in this study could provide a foundational framework for the conception and assessment of innovative strategies targeted at enhancing zero-shot multimodal techniques. The evaluation pipeline and benchmark are available at //github.com/Yuliang-Liu/MultimodalOCR.
Characters do not convey meaning, but sequences of characters do. We propose an unsupervised distributional method to learn the abstract meaningful units in a sequence of characters. Rather than segmenting the sequence, our Dynamic Capacity Slot Attention model discovers continuous representations of the objects in the sequence, extending an architecture for object discovery in images. We train our model on different languages and evaluate the quality of the obtained representations with forward and reverse probing classifiers. These experiments show that our model succeeds in discovering units which are similar to those proposed previously in form, content and level of abstraction, and which show promise for capturing meaningful information at a higher level of abstraction.
It is often challenging for a user to articulate their preferences accurately in multi-objective decision-making problems. Demonstration-based preference inference (DemoPI) is a promising approach to mitigate this problem. Understanding the behaviours and values of energy customers is an example of a scenario where preference inference can be used to gain insights into the values of energy customers with multiple objectives, e.g. cost and comfort. In this work, we applied the state-of-art DemoPI method, i.e., the dynamic weight-based preference inference (DWPI) algorithm in a multi-objective residential energy consumption setting to infer preferences from energy consumption demonstrations by simulated users following a rule-based approach. According to our experimental results, the DWPI model achieves accurate demonstration-based preference inferring in three scenarios. These advancements enhance the usability and effectiveness of multi-objective reinforcement learning (MORL) in energy management, enabling more intuitive and user-friendly preference specifications, and opening the door for DWPI to be applied in real-world settings.
Dynamic mode decomposition (DMD) has recently become a popular tool for the non-intrusive analysis of dynamical systems. Exploiting Proper Orthogonal Decomposition (POD) as a dimensionality reduction technique, DMD is able to approximate a dynamical system as a sum of spatial basis evolving linearly in time, thus enabling a better understanding of the physical phenomena and forecasting of future time instants. In this work we propose an extension of DMD to parameterized dynamical systems, focusing on the future forecasting of the output of interest in a parametric context. Initially all the snapshots -- for different parameters and different time instants -- are projected to a reduced space; then DMD, or one of its variants, is employed to approximate reduced snapshots for future time instants. Exploiting the low dimension of the reduced space the predicted reduced snapshots are then combined using regression techniques, thus enabling the possibility to approximate any untested parametric configuration in future. This paper depicts in detail the algorithmic core of this method; we also present and discuss three test cases for our algorithm: a simple dynamical system with a linear parameter dependency, a heat problem with nonlinear parameter dependency and a fluid dynamics problem with nonlinear parameter dependency.
Memory corruption is an important class of vulnerability that can be leveraged to craft control flow hijacking attacks. Control Flow Integrity (CFI) provides protection against such attacks. Application of type-based CFI policies requires information regarding the number and type of function arguments. Binary-level type recovery is inherently speculative, which motivates the need for an evaluation framework to assess the effectiveness of binary-level CFI techniques compared with their source-level counterparts, where such type information is fully and accurately accessible. In this work, we develop a novel, generalized and extensible framework to assess how the program analysis information we get from state-of-the-art binary analysis tools affects the efficacy of type-based CFI techniques. We introduce new and insightful metrics to quantitatively compare source independent CFI policies with their ground truth source aware counterparts. We leverage our framework to evaluate binary-level CFI policies implemented using program analysis information extracted from the IDA Pro binary analyzer and compared with the ground truth information obtained from the LLVM compiler, and present our observations.
Users develop mental models of robots to conceptualize what kind of interactions they can have with those robots. The conceptualizations are often formed before interactions with the robot and are based only on observing the robot's physical design. As a result, understanding conceptualizations formed from physical design is necessary to understand how users intend to interact with the robot. We propose to use multimodal features of robot embodiments to predict what kinds of expectations users will have about a given robot's social and physical capabilities. We show that using such features provides information about general mental models of the robots that generalize across socially interactive robots. We describe how these models can be incorporated into interaction design and physical design for researchers working with socially interactive robots.
The problem is considered of optimizing discrete parameters in the presence of constraints. We use the stochastic sigmoid with temperature and put forward the new adaptive gradient method CONGA. The search for an optimal solution is carried out by a population of individuals. Each of them varies according to gradients of the 'environment' and is characterized by two temperature parameters with different annealing schedules. Unadapted individuals die, and optimal ones interbreed, the result is directed evolutionary dynamics. The proposed method is illustrated using the well-known combinatorial problem for optimal packing of a backpack (0-1 KP).
Recent investigations demonstrate that adversarial patches can be utilized to manipulate the result of object detection models. However, the conspicuous patterns on these patches may draw more attention and raise suspicions among humans. Moreover, existing works have primarily focused on enhancing the efficacy of attacks in the physical domain, rather than seeking to optimize their stealth attributes and transferability potential. To address these issues, we introduce a dual-perception-based attack framework that generates an adversarial patch known as the More Vivid Patch (MVPatch). The framework consists of a model-perception degradation method and a human-perception improvement method. To derive the MVPatch, we formulate an iterative process that simultaneously constrains the efficacy of multiple object detectors and refines the visual correlation between the generated adversarial patch and a realistic image. Our method employs a model-perception-based approach that reduces the object confidence scores of several object detectors to boost the transferability of adversarial patches. Further, within the human-perception-based framework, we put forward a lightweight technique for visual similarity measurement that facilitates the development of inconspicuous and natural adversarial patches and eliminates the reliance on additional generative models. Additionally, we introduce the naturalness score and transferability score as metrics for an unbiased assessment of various adversarial patches' natural appearance and transferability capacity. Extensive experiments demonstrate that the proposed MVPatch algorithm achieves superior attack transferability compared to similar algorithms in both digital and physical domains while also exhibiting a more natural appearance. These findings emphasize the remarkable stealthiness and transferability of the proposed MVPatch attack algorithm.
Although large language models (LLMs) are impressive in solving various tasks, they can quickly be outdated after deployment. Maintaining their up-to-date status is a pressing concern in the current era. This paper provides a comprehensive review of recent advances in aligning LLMs with the ever-changing world knowledge without re-training from scratch. We categorize research works systemically and provide in-depth comparisons and discussion. We also discuss existing challenges and highlight future directions to facilitate research in this field. We release the paper list at //github.com/hyintell/awesome-refreshing-llms
Solving complicated AI tasks with different domains and modalities is a key step toward artificial general intelligence. While there are abundant AI models available for different domains and modalities, they cannot handle complicated AI tasks. Considering large language models (LLMs) have exhibited exceptional ability in language understanding, generation, interaction, and reasoning, we advocate that LLMs could act as a controller to manage existing AI models to solve complicated AI tasks and language could be a generic interface to empower this. Based on this philosophy, we present HuggingGPT, a framework that leverages LLMs (e.g., ChatGPT) to connect various AI models in machine learning communities (e.g., Hugging Face) to solve AI tasks. Specifically, we use ChatGPT to conduct task planning when receiving a user request, select models according to their function descriptions available in Hugging Face, execute each subtask with the selected AI model, and summarize the response according to the execution results. By leveraging the strong language capability of ChatGPT and abundant AI models in Hugging Face, HuggingGPT is able to cover numerous sophisticated AI tasks in different modalities and domains and achieve impressive results in language, vision, speech, and other challenging tasks, which paves a new way towards artificial general intelligence.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191