Pretrained visual-language models have made significant advancements in multimodal tasks, including image-text retrieval. However, a major challenge in image-text matching lies in language bias, where models predominantly rely on language priors and neglect to adequately consider the visual content. We thus present Multimodal ASsociation Score (MASS), a framework that reduces the reliance on language priors for better visual accuracy in image-text matching problems. It can be seamlessly incorporated into existing visual-language models without necessitating additional training. Our experiments have shown that MASS effectively lessens language bias without losing an understanding of linguistic compositionality. Overall, MASS offers a promising solution for enhancing image-text matching performance in visual-language models.
The field of medical diagnostics has witnessed a transformative convergence of artificial intelligence (AI) and healthcare data, offering promising avenues for enhancing patient care and disease comprehension. However, this integration of multimodal data, specifically histopathology whole slide images (WSIs) and genetic sequencing data, presents unique challenges due to modality disparities and the need for scalable computational solutions. This paper addresses the scarcity of multimodal solutions, primarily centered around unimodal data solutions, thus limiting the realization of the rich insights that can be derived from integrating images and genomic data. Here, we introduce MarbliX ``Multimodal Association and Retrieval with Binary Latent Indexed matriX,'' an innovative multimodal framework that integrates histopathology images with immunogenomic sequencing data, encapsulating them into a concise binary patient code, referred to as ``monogram.'' This binary representation facilitates the establishment of a comprehensive archive, enabling clinicians to match similar cases. The experimental results demonstrate the potential of MarbliX to empower healthcare professionals with in-depth insights, leading to more precise diagnoses, reduced variability, and expanded personalized treatment options, particularly in the context of cancer.
In this paper, we present Vi-Fi, a multi-modal system that leverages a user’s smartphone WiFi Fine Timing Measurements (FTM) and inertial measurement unit (IMU) sensor data to associate the user detected on a camera footage with their corresponding smartphone identifier (e.g. WiFi MAC address). Our approach uses a recurrent multi-modal deep neural network that exploits FTM and IMU measurements along with distance between user and camera (depth information) to learn affinity matrices. As a baseline method for comparison, we also present a traditional non deep learning approach that uses bipartite graph matching. To facilitate evaluation, we collected a multi-modal dataset that comprises camera videos with depth information (RGB-D), WiFi FTM and IMU measurements for multiple participants at diverse real-world settings. Using association accuracy as the key metric for evaluating the fidelity of 𝑉𝑖𝑠𝑢𝑎𝑙4 𝑉𝑖𝑠𝑢𝑎𝑙5 Figure 1: Motivation: Successfully associating vision-wireless Vi-Fi in associating human users on camera feed with their phone IDs, we show that Vi-Fi achieves between 81% (real-time) to 91% (offline) association accuracy.
While vision-and-language models perform well on tasks such as visual question answering, they struggle when it comes to basic human commonsense reasoning skills. In this work, we introduce WinoGAViL: an online game of vision-and-language associations (e.g., between werewolves and a full moon), used as a dynamic evaluation benchmark. Inspired by the popular card game Codenames, a spymaster gives a textual cue related to several visual candidates, and another player tries to identify them. Human players are rewarded for creating associations that are challenging for a rival AI model but still solvable by other human players. We use the game to collect 3.5K instances, finding that they are intuitive for humans (>90% Jaccard index) but challenging for state-of-the-art AI models, where the best model (ViLT) achieves a score of 52%, succeeding mostly where the cue is visually salient. Our analysis as well as the feedback we collect from players indicate that the collected associations require diverse reasoning skills, including general knowledge, common sense, abstraction, and more. We release the dataset, the code and the interactive game, allowing future data collection that can be used to develop models with better association abilities.
The field of artificial intelligence has significantly advanced over the past decades, inspired by discoveries from the fields of biology and neuroscience. The idea of this work is inspired by the process of self-organization of cortical areas in the human brain from both afferent and lateral/internal connections. In this work, we develop an original brain-inspired neural model associating Self-Organizing Maps (SOM) and Hebbian learning in the Reentrant SOM (ReSOM) model. The framework is applied to multimodal classification problems. Compared to existing methods based on unsupervised learning with post-labeling, the model enhances the state-of-the-art results. This work also demonstrates the distributed and scalable nature of the model through both simulation results and hardware execution on a dedicated FPGA-based platform named SCALP (Self-configurable 3D Cellular Adaptive Platform). SCALP boards can be interconnected in a modular way to support the structure of the neural model. Such a unified software and hardware approach enables the processing to be scaled and allows information from several modalities to be merged dynamically. The deployment on hardware boards provides performance results of parallel execution on several devices, with the communication between each board through dedicated serial links. The proposed unified architecture, composed of the ReSOM model and the SCALP hardware platform, demonstrates a significant increase in accuracy thanks to multimodal association, and a good trade-off between latency and power consumption compared to a centralized GPU implementation.
Cortical plasticity is one of the main features that enable our ability to learn and adapt in our environment. Indeed, the cerebral cortex self-organizes itself through structural and synaptic plasticity mechanisms that are very likely at the basis of an extremely interesting characteristic of the human brain development: the multimodal association. In spite of the diversity of the sensory modalities, like sight, sound and touch, the brain arrives at the same concepts (convergence). Moreover, biological observations show that one modality can activate the internal representation of another modality when both are correlated (divergence). In this work, we propose the Reentrant Self-Organizing Map (ReSOM), a brain-inspired neural system based on the reentry theory using Self-Organizing Maps and Hebbian-like learning. We propose and compare different computational methods for unsupervised learning and inference, then quantify the gain of the ReSOM in a multimodal classification task. The divergence mechanism is used to label one modality based on the other, while the convergence mechanism is used to improve the overall accuracy of the system. We perform our experiments on a constructed written/spoken digits database and a DVS/EMG hand gestures database. The proposed model is implemented on a cellular neuromorphic architecture that enables distributed computing with local connectivity. We show the gain of the so-called hardware plasticity induced by the ReSOM, where the system's topology is not fixed by the user but learned along the system's experience through self-organization.