Cued Speech (CS) is an advanced visual phonetic encoding system that integrates lip reading with hand codings, enabling people with hearing impairments to communicate efficiently. CS video generation aims to produce specific lip and gesture movements of CS from audio or text inputs. The main challenge is that given limited CS data, we strive to simultaneously generate fine-grained hand and finger movements, as well as lip movements, meanwhile the two kinds of movements need to be asynchronously aligned. Existing CS generation methods are fragile and prone to poor performance due to template-based statistical models and careful hand-crafted pre-processing to fit the models. Therefore, we propose a novel Gloss-prompted Diffusion-based CS Gesture generation framework (called GlossDiff). Specifically, to integrate additional linguistic rules knowledge into the model. we first introduce a bridging instruction called \textbf{Gloss}, which is an automatically generated descriptive text to establish a direct and more delicate semantic connection between spoken language and CS gestures. Moreover, we first suggest rhythm is an important paralinguistic feature for CS to improve the communication efficacy. Therefore, we propose a novel Audio-driven Rhythmic Module (ARM) to learn rhythm that matches audio speech. Moreover, in this work, we design, record, and publish the first Chinese CS dataset with four CS cuers. Extensive experiments demonstrate that our method quantitatively and qualitatively outperforms current state-of-the-art (SOTA) methods. We release the code and data at https://glossdiff.github.io/.
Linguistic ambiguity continues to represent a significant challenge for natural language processing (NLP) systems, notwithstanding the advancements in architectures such as Transformers and BERT. Inspired by the recent success of instructional models like ChatGPT and Gemini (In 2023, the artificial intelligence was called Bard.), this study aims to analyze and discuss linguistic ambiguity within these models, focusing on three types prevalent in Brazilian Portuguese: semantic, syntactic, and lexical ambiguity. We create a corpus comprising 120 sentences, both ambiguous and unambiguous, for classification, explanation, and disambiguation. The models capability to generate ambiguous sentences was also explored by soliciting sets of sentences for each type of ambiguity. The results underwent qualitative analysis, drawing on recognized linguistic references, and quantitative assessment based on the accuracy of the responses obtained. It was evidenced that even the most sophisticated models, such as ChatGPT and Gemini, exhibit errors and deficiencies in their responses, with explanations often providing inconsistent. Furthermore, the accuracy peaked at 49.58 percent, indicating the need for descriptive studies for supervised learning.
In high-pressure environments where human individuals must simultaneously monitor multiple entities, communicate effectively, and maintain intense focus, the perception of time becomes a critical factor influencing performance and well-being. One indicator of well-being can be the person's subjective time perception. In our project $ChronoPilot$, we aim to develop a device that modulates human subjective time perception. In this study, we present a method to automatically assess the subjective time perception of air traffic controllers, a group often faced with demanding conditions, using their physiological data and eleven state-of-the-art machine learning classifiers. The physiological data consist of photoplethysmogram, electrodermal activity, and temperature data. We find that the support vector classifier works best with an accuracy of 79 % and electrodermal activity provides the most descriptive biomarker. These findings are an important step towards closing the feedback loop of our $ChronoPilot$-device to automatically modulate the user's subjective time perception. This technological advancement may promise improvements in task management, stress reduction, and overall productivity in high-stakes professions.
We study interactive learning of LLM-based language agents based on user edits made to the agent's output. In a typical setting such as writing assistants, the user interacts with a language agent to generate a response given a context, and may optionally edit the agent response to personalize it based on their latent preference, in addition to improving the correctness. The edit feedback is naturally generated, making it a suitable candidate for improving the agent's alignment with the user's preference, and for reducing the cost of user edits over time. We propose a learning framework, PRELUDE that infers a description of the user's latent preference based on historic edit data. The inferred user preference descriptions are used to define prompts for generating responses in the future. This avoids fine-tuning the agent, which is costly, challenging to scale with the number of users, and may even degrade its performance on other tasks. Furthermore, learning descriptive preference improves interpretability, allowing the user to view and modify the learned preference. However, user preference can be complex, subtle, and vary based on context, making it challenging to learn. To address this, we propose a simple yet effective algorithm named CIPHER that leverages the LLM to infer the user preference for a given context based on user edits. In the future, CIPHER retrieves inferred preferences from the k-closest contexts in the history, and forms an aggregate preference for response generation. We introduce two interactive environments -- summarization and email writing, and use a GPT-4 simulated user for evaluation. On both tasks, CIPHER outperforms several baselines by achieving the lowest edit distance cost while only having a small overhead in LLM query cost. Our analysis reports that user preferences learned by CIPHER show significant similarity to the ground truth latent preferences.
This paper presents the second part of the two-part survey series on decomposition-based evolutionary multi-objective optimization where we mainly focus on discussing the literature related to multi-objective evolutionary algorithms based on decomposition (MOEA/D). Complementary to the first part, here we employ a series of advanced data mining approaches to provide a comprehensive anatomy of the enormous landscape of MOEA/D research, which is far beyond the capacity of classic manual literature review protocol. In doing so, we construct a heterogeneous knowledge graph that encapsulates more than 5,400 papers, 10,000 authors, 400 venues, and 1,600 institutions for MOEA/D research. We start our analysis with basic descriptive statistics. Then we delve into prominent research/application topics pertaining to MOEA/D with state-of-the-art topic modeling techniques and interrogate their sptial-temporal and bilateral relationships. We also explored the collaboration and citation networks of MOEA/D, uncovering hidden patterns in the growth of literature as well as collaboration between researchers. Our data mining results here, combined with the expert review in Part I, together offer a holistic view of the MOEA/D research, and demonstrate the potential of an exciting new paradigm for conducting scientific surveys from a data science perspective.
With the digital imagery landscape rapidly evolving, image stocks and AI-generated image marketplaces have become central to visual media. Traditional stock images now exist alongside innovative platforms that trade in prompts for AI-generated visuals, driven by sophisticated APIs like DALL-E 3 and Midjourney. This paper studies the possibility of employing multi-modal models with enhanced visual understanding to mimic the outputs of these platforms, introducing an original attack strategy. Our method leverages fine-tuned CLIP models, a multi-label classifier, and the descriptive capabilities of GPT-4V to create prompts that generate images similar to those available in marketplaces and from premium stock image providers, yet at a markedly lower expense. In presenting this strategy, we aim to spotlight a new class of economic and security considerations within the realm of digital imagery. Our findings, supported by both automated metrics and human assessment, reveal that comparable visual content can be produced for a fraction of the prevailing market prices ($0.23 - $0.27 per image), emphasizing the need for awareness and strategic discussions about the integrity of digital media in an increasingly AI-integrated landscape. Our work also contributes to the field by assembling a dataset consisting of approximately 19 million prompt-image pairs generated by the popular Midjourney platform, which we plan to release publicly.
Text-to-Image (T2I) Synthesis has made tremendous strides in enhancing synthesized image quality, but current datasets evaluate model performance only on descriptive, instruction-based prompts. Real-world news image captions take a more pragmatic approach, providing high-level situational and Named-Entity (NE) information and limited physical object descriptions, making them abstractive. To evaluate the ability of T2I models to capture intended subjects from news captions, we introduce the Abstractive News Captions with High-level cOntext Representation (ANCHOR) dataset, containing 70K+ samples sourced from 5 different news media organizations. With Large Language Models (LLM) achieving success in language and commonsense reasoning tasks, we explore the ability of different LLMs to identify and understand key subjects from abstractive captions. Our proposed method Subject-Aware Finetuning (SAFE), selects and enhances the representation of key subjects in synthesized images by leveraging LLM-generated subject weights. It also adapts to the domain distribution of news images and captions through custom Domain Fine-tuning, outperforming current T2I baselines on ANCHOR. By launching the ANCHOR dataset, we hope to motivate research in furthering the Natural Language Understanding (NLU) capabilities of T2I models.
Multi-modal knowledge graph completion (MMKGC) aims to discover unobserved knowledge from given knowledge graphs, collaboratively leveraging structural information from the triples and multi-modal information of the entities to overcome the inherent incompleteness. Existing MMKGC methods usually extract multi-modal features with pre-trained models, resulting in coarse handling of multi-modal entity information, overlooking the nuanced, fine-grained semantic details and their complex interactions. To tackle this shortfall, we introduce a novel framework MyGO to tokenize, fuse, and augment the fine-grained multi-modal representations of entities and enhance the MMKGC performance. Motivated by the tokenization technology, MyGO tokenizes multi-modal entity information as fine-grained discrete tokens and learns entity representations with a cross-modal entity encoder. To further augment the multi-modal representations, MyGO incorporates fine-grained contrastive learning to highlight the specificity of the entity representations. Experiments on standard MMKGC benchmarks reveal that our method surpasses 19 of the latest models, underlining its superior performance. Code and data can be found in https://github.com/zjukg/MyGO
Traffic video description and analysis have received much attention recently due to the growing demand for efficient and reliable urban surveillance systems. Most existing methods only focus on locating traffic event segments, which severely lack descriptive details related to the behaviour and context of all the subjects of interest in the events. In this paper, we present TrafficVLM, a novel multi-modal dense video captioning model for vehicle ego camera view. TrafficVLM models traffic video events at different levels of analysis, both spatially and temporally, and generates long fine-grained descriptions for the vehicle and pedestrian at different phases of the event. We also propose a conditional component for TrafficVLM to control the generation outputs and a multi-task fine-tuning paradigm to enhance TrafficVLM's learning capability. Experiments show that TrafficVLM performs well on both vehicle and overhead camera views. Our solution achieved outstanding results in Track 2 of the AI City Challenge 2024, ranking us third in the challenge standings. Our code is publicly available at https://github.com/quangminhdinh/TrafficVLM.
Concept Bottleneck Models (CBMs) map the black-box visual representations extracted by deep neural networks onto a set of interpretable concepts and use the concepts to make predictions, enhancing the transparency of the decision-making process. Multimodal pre-trained models can match visual representations with textual concept embeddings, allowing for obtaining the interpretable concept bottleneck without the expertise concept annotations. Recent research has focused on the concept bank establishment and the high-quality concept selection. However, it is challenging to construct a comprehensive concept bank through humans or large language models, which severely limits the performance of CBMs. In this work, we propose the Incremental Residual Concept Bottleneck Model (Res-CBM) to address the challenge of concept completeness. Specifically, the residual concept bottleneck model employs a set of optimizable vectors to complete missing concepts, then the incremental concept discovery module converts the complemented vectors with unclear meanings into potential concepts in the candidate concept bank. Our approach can be applied to any user-defined concept bank, as a post-hoc processing method to enhance the performance of any CBMs. Furthermore, to measure the descriptive efficiency of CBMs, the Concept Utilization Efficiency (CUE) metric is proposed. Experiments show that the Res-CBM outperforms the current state-of-the-art methods in terms of both accuracy and efficiency and achieves comparable performance to black-box models across multiple datasets.
Search engines these days can serve datasets as search results. Datasets get picked up by search technologies based on structured descriptions on their official web pages, informed by metadata ontologies such as the Dataset content type of schema.org. Despite this promotion of the content type dataset as a first-class citizen of search results, a vast proportion of datasets, particularly research datasets, still need to be made discoverable and, therefore, largely remain unused. This is due to the sheer volume of datasets released every day and the inability of metadata to reflect a dataset's content and context accurately. This work seeks to improve this situation for a specific class of datasets, namely research datasets, which are the result of research endeavors and are accompanied by a scholarly publication. We propose the ORKG-Dataset content type, a specialized branch of the Open Research Knowledge Graoh (ORKG) platform, which provides descriptive information and a semantic model for research datasets, integrating them with their accompanying scholarly publications. This work aims to establish a standardized framework for recording and reporting research datasets within the ORKG-Dataset content type. This, in turn, increases research dataset transparency on the web for their improved discoverability and applied use. In this paper, we present a proposal -- the minimum FAIR, comparable, semantic description of research datasets in terms of salient properties of their supporting publication. We design a specific application of the ORKG-Dataset semantic model based on 40 diverse research datasets on scientific information extraction.
Integration of Large Language Models (LLMs) into visual domain tasks, resulting in visual-LLMs (V-LLMs), has enabled exceptional performance in vision-language tasks, particularly for visual question answering (VQA). However, existing V-LLMs (e.g. BLIP-2, LLaVA) demonstrate weak spatial reasoning and localization awareness. Despite generating highly descriptive and elaborate textual answers, these models fail at simple tasks like distinguishing a left vs right location. In this work, we explore how image-space coordinate based instruction fine-tuning objectives could inject spatial awareness into V-LLMs. We discover optimal coordinate representations, data-efficient instruction fine-tuning objectives, and pseudo-data generation strategies that lead to improved spatial awareness in V-LLMs. Additionally, our resulting model improves VQA across image and video domains, reduces undesired hallucination, and generates better contextual object descriptions. Experiments across 5 vision-language tasks involving 14 different datasets establish the clear performance improvements achieved by our proposed framework.
Evaluating Large Language Models (LLMs) is challenging due to their generative nature, necessitating precise evaluation methodologies. Additionally, non-English LLM evaluation lags behind English, resulting in the absence or weakness of LLMs for many languages. In response to this necessity, we introduce Khayyam Challenge (also known as PersianMMLU), a meticulously curated collection comprising 20,192 four-choice questions sourced from 38 diverse tasks extracted from Persian examinations, spanning a wide spectrum of subjects, complexities, and ages. The primary objective of the Khayyam Challenge is to facilitate the rigorous evaluation of LLMs that support the Persian language. Distinctive features of the Khayyam Challenge are (i) its comprehensive coverage of various topics, including literary comprehension, mathematics, sciences, logic, intelligence testing, etc., aimed at assessing different facets of LLMs such as language comprehension, reasoning, and information retrieval across various educational stages, from lower primary school to upper secondary school (ii) its inclusion of rich metadata such as human response rates, difficulty levels, and descriptive answers (iii) its utilization of new data to avoid data contamination issues prevalent in existing frameworks (iv) its use of original, non-translated data tailored for Persian speakers, ensuring the framework is free from translation challenges and errors while encompassing cultural nuances (v) its inherent scalability for future data updates and evaluations without requiring special human effort. Previous works lacked an evaluation framework that combined all of these features into a single comprehensive benchmark. Furthermore, we evaluate a wide range of existing LLMs that support the Persian language, with statistical analyses and interpretations of their outputs.
Condition monitoring of industrial systems is crucial for ensuring safety and maintenance planning, yet notable challenges arise in real-world settings due to the limited or non-existent availability of fault samples. This paper introduces an innovative solution to this problem by proposing a new method for fault detection and condition monitoring for unseen data. Adopting an approach inspired by zero-shot learning, our method can identify faults and assign a relative health index to various operational conditions. Typically, we have plenty of data on normal operations, some data on compromised conditions, and very few (if any) samples of severe faults. We use a variational autoencoder to capture the probabilistic distribution of previously seen and new unseen conditions. The health status is determined by comparing each sample's deviation from a normal operation reference distribution in the latent space. Faults are detected by establishing a threshold for the health indexes, allowing the model to identify severe, unseen faults with high accuracy, even amidst noise. We validate our approach using the run-to-failure IMS-bearing dataset and compare it with other methods. The health indexes generated by our model closely match the established descriptive model of bearing wear, attesting to the robustness and reliability of our method. These findings highlight the potential of our methodology in augmenting fault detection capabilities within industrial domains, thereby contributing to heightened safety protocols and optimized maintenance practices.
Despite significant advancements, large multimodal models (LMMs) still struggle to bridge the gap between low-level visual perception -- focusing on shapes, sizes, and layouts -- and high-level language reasoning, such as semantics and logic. This limitation is evident in tasks that require precise visual perception, like comparing geometric properties or solving visual reasoning problems. To study this failure mode, we focus on vector graphics -- images composed of 2D objects and shapes, prevalent in LMM-based tasks in web, design, and OS environments. We identify two key research questions: how can we enable precise visual perception, and how can we facilitate high-level reasoning based on such low-level perceptions? To capture fine visual details, we use Scalable Vector Graphics (SVG) for accurate encoding of visual scenes. However, SVGs are not readily interpretable by LMMs in a zero-shot manner. To tackle this, we propose the Visually Descriptive Language Model (VDLM), which introduces a Primal Visual Description (PVD) as an intermediate textual representation. PVD translates SVGs into a text-based abstraction consisting of primitive attributes (e.g., shape, position, measurement) and their corresponding values. PVD can be learned using task-agnostic synthesized data and represents visual primitives that are universal across vector graphics. This abstraction is more structured, allowing for direct interpretation by foundation models for zero-shot generalization. Without human-annotated data, empirical results show that VDLM significantly improves state-of-the-art LMMs like GPT-4o on various multimodal perception and reasoning tasks. Extensive analyses of VDLM show improved interpretability due to its disentangled perception and reasoning. We also demonstrate a positive correlation between PVD quality and task performance. Project page: https://mikewangwzhl.github.io/VDLM/
This paper describes our participation in the Shared Task on Software Mentions Disambiguation (SOMD), with a focus on improving relation extraction in scholarly texts through generative Large Language Models (LLMs) using single-choice question-answering. The methodology prioritises the use of in-context learning capabilities of GLMs to extract software-related entities and their descriptive attributes, such as distributive information. Our approach uses Retrieval-Augmented Generation (RAG) techniques and GLMs for Named Entity Recognition (NER) and Attributive NER to identify relationships between extracted software entities, providing a structured solution for analysing software citations in academic literature. The paper provides a detailed description of our approach, demonstrating how using GLMs in a single-choice QA paradigm can greatly enhance IE methodologies. Our participation in the SOMD shared task highlights the importance of precise software citation practices and showcases our system's ability to overcome the challenges of disambiguating and extracting relationships between software mentions. This sets the groundwork for future research and development in this field.
Generative models have shown a giant leap in synthesizing photo-realistic images with minimal expertise, sparking concerns about the authenticity of online information. This study aims to develop a universal AI-generated image detector capable of identifying images from diverse sources. Existing methods struggle to generalize across unseen generative models when provided with limited sample sources. Inspired by the zero-shot transferability of pre-trained vision-language models, we seek to harness the nontrivial visual-world knowledge and descriptive proficiency of CLIP-ViT to generalize over unknown domains. This paper presents a novel parameter-efficient fine-tuning approach, mixture of low-rank experts, to fully exploit CLIP-ViT's potential while preserving knowledge and expanding capacity for transferable detection. We adapt only the MLP layers of deeper ViT blocks via an integration of shared and separate LoRAs within an MoE-based structure. Extensive experiments on public benchmarks show that our method achieves superiority over state-of-the-art approaches in cross-generator generalization and robustness to perturbations. Remarkably, our best-performing ViT-L/14 variant requires training only 0.08% of its parameters to surpass the leading baseline by +3.64% mAP and +12.72% avg.Acc across unseen diffusion and autoregressive models. This even outperforms the baseline with just 0.28% of the training data. Our code and pre-trained models will be available at https://github.com/zhliuworks/CLIPMoLE.
We propose the new task 'open-world video instance segmentation and captioning'. It requires to detect, segment, track and describe with rich captions never before seen objects. This challenging task can be addressed by developing "abstractors" which connect a vision model and a language foundation model. Concretely, we connect a multi-scale visual feature extractor and a large language model (LLM) by developing an object abstractor and an object-to-text abstractor. The object abstractor, consisting of a prompt encoder and transformer blocks, introduces spatially-diverse open-world object queries to discover never before seen objects in videos. An inter-query contrastive loss further encourages the diversity of object queries. The object-to-text abstractor is augmented with masked cross-attention and acts as a bridge between the object queries and a frozen LLM to generate rich and descriptive object-centric captions for each detected object. Our generalized approach surpasses the baseline that jointly addresses the tasks of open-world video instance segmentation and dense video object captioning by 13% on never before seen objects, and by 10% on object-centric captions.
To enhance the interpretability of Reinforcement Learning (RL), we propose Revealing Evolutionary Action Consequence Trajectories (REACT). In contrast to the prevalent practice of validating RL models based on their optimal behavior learned during training, we posit that considering a range of edge-case trajectories provides a more comprehensive understanding of their inherent behavior. To induce such scenarios, we introduce a disturbance to the initial state, optimizing it through an evolutionary algorithm to generate a diverse population of demonstrations. To evaluate the fitness of trajectories, REACT incorporates a joint fitness function that encourages both local and global diversity in the encountered states and chosen actions. Through assessments with policies trained for varying durations in discrete and continuous environments, we demonstrate the descriptive power of REACT. Our results highlight its effectiveness in revealing nuanced aspects of RL models' behavior beyond optimal performance, thereby contributing to improved interpretability.
We propose VecKM, a local point cloud geometry encoder that is descriptive and efficient to compute. VecKM leverages a unique approach by vectorizing a kernel mixture to represent the local point cloud. Such representation's descriptiveness is supported by two theorems that validate its ability to reconstruct and preserve the similarity of the local shape. Unlike existing encoders downsampling the local point cloud, VecKM constructs the local geometry encoding using all neighboring points, producing a more descriptive encoding. Moreover, VecKM is efficient to compute and scalable to large point cloud inputs: VecKM reduces the memory cost from $(n^2+nKd)$ to $(nd+np)$; and reduces the major runtime cost from computing $nK$ MLPs to $n$ MLPs, where $n$ is the size of the point cloud, $K$ is the neighborhood size, $d$ is the encoding dimension, and $p$ is a marginal factor. The efficiency is due to VecKM's unique factorizable property that eliminates the need of explicitly grouping points into neighbors. In the normal estimation task, VecKM demonstrates not only 100x faster inference speed but also highest accuracy and strongest robustness. In classification and segmentation tasks, integrating VecKM as a preprocessing module achieves consistently better performance than the PointNet, PointNet++, and point transformer baselines, and runs consistently faster by up to 10 times.