More Like this

1. AudiFace: Multimodal Deep Learning for Depression Screening

   Ricardo Flores, ML Tlachac (December 2022, Proceedings of Machine Learning Research)

   Depression is a very common mental health disorder with a devastating social and economic impact. It can be costly and difficult to detect, traditionally requiring a significant number of hours by a trained mental health professional. Recently, machine learning and deep learning models have been trained for depression screening using modalities extracted from videos of clinical interviews conducted by a virtual agent. This complex task is challenging for deep learning models because of the multiple modalities and limited number of participants in the dataset. To address these challenges we propose AudiFace, a multimodal deep learning model that inputs temporal facial features, audio, and transcripts to screen for depression. To incorporate all three modalities, AudiFace combines multiple pre-trained transfer learning models and bidirectional LSTM with self-Attention. When compared with the state-of-the-art models, AudiFace achieves the highest F1 scores for thirteen of the fifteen different datasets. AudiFace notably improves the depression screening capabilities of general wellbeing questions. Eye gaze proved to be the most valuable of the temporal facial features, both in the unimodal and multimodal models. Our results can be used to determine the best combination of modalities, temporal facial features, as well as clinical interview questions for future depression screening applications. 

   more » « less
2. What do MLLMs hear? Examining reasoning with text and audio encoder components in Multimodal Large Language Models.

   Coban, Enis Berk; Mandel, Michael I; Devaney, Johanna (October 2024, NeurIPS Audio Imagination Workshop)

   Large Language Models (LLMs) have demonstrated remarkable reasoning capabilities, notably in connecting ideas and adhering to logical rules to solve problems. These models have evolved to accommodate various data modalities, including sound and images, known as multimodal LLMs (MLLMs), which are capable of describing images or sound recordings. Previous work has demonstrated that when the LLM component in MLLMs is frozen, the audio or visual encoder serves to caption the sound or image input facilitating text-based reasoning with the LLM component. We are interested in using the LLM's reasoning capabilities in order to facilitate classification. In this paper, we demonstrate through a captioning/classification experiment that an audio MLLM cannot fully leverage its LLM's text-based reasoning when generating audio captions. We also consider how this may be due to MLLMs separately representing auditory and textual information such that it severs the reasoning pathway from the LLM to the audio encoder. 

   more » « less
3. Improving Speech Emotion Recognition Using Self-Supervised Learning with Domain-Specific Audiovisual Tasks

   https://doi.org/10.21437/Interspeech.2022-11012  

   Goncalves, Lucas; Busso, Carlos (September 2022, Interspeech 2022)

   Speech emotion recognition (SER) is a challenging task due to the limited availability of real-world labeled datasets. Since it is easier to find unlabeled data, the use of self-supervised learning (SSL) has become an attractive alternative. This study proposes new pre-text tasks for SSL to improve SER. While our target application is SER, the proposed pre-text tasks include audio-visual formulations, leveraging the relationship between acoustic and facial features. Our proposed approach introduces three new unimodal and multimodal pre-text tasks that are carefully designed to learn better representations for predicting emotional cues from speech. Task 1 predicts energy variations (high or low) from a speech sequence. Task 2 uses speech features to predict facial activation (high or low) based on facial landmark movements. Task 3 performs a multi-class emotion recognition task on emotional labels obtained from combinations of action units (AUs) detected across a video sequence. We pre-train a network with 60.92 hours of unlabeled data, fine-tuning the model for the downstream SER task. The results on the CREMA-D dataset show that the model pre-trained on the proposed domain-specific pre-text tasks significantly improves the precision (up to 5.1%), recall (up to 4.5%), and F1-scores (up to 4.9%) of our SER system. 

   more » « less
4. Protecting Privacy in Multimodal Large Language Models with MLLMU-Bench

   Liu, Zheyuan; Dou, Guangyao; Jia, Mengzhao; Tan, Zhaoxuan; Zeng, Qingkai; Yuan, Yongle; Jiang, Meng (April 2025, Association for Computational Linguistics)

   Generative models such as Large Language Models (LLM) and Multimodal Large Language models (MLLMs) trained on massive web corpora can memorize and disclose individuals’ confidential and private data, raising legal and ethical concerns. While many previous works have addressed this issue in LLM via machine unlearning, it remains largely unexplored for MLLMs. To tackle this challenge, we introduce Multimodal Large Language Model Unlearning Benchmark (MLLMU-Bench), a novel benchmark aimed at advancing the understanding of multimodal machine unlearning. MLLMU-Bench consists of 500 fictitious profiles and 153 profiles for public celebrities, each profile feature over 14 customized question-answer pairs, evaluated from both multimodal (image+text) and unimodal (text) perspectives. The benchmark is divided into four sets to assess unlearning algorithms in terms of efficacy, generalizability, and model utility. Finally, we provide baseline results using existing generative model unlearning algorithms. Surprisingly, our experiments show that unimodal unlearning algorithms excel in generation tasks, while multimodal unlearning approaches perform better in classification with multimodal inputs. 

   more » « less
5. Coarse-to-Fine Vision-Language Pre-training with Fusion in the Backbone

   Dou, Zi-Yi; Kamath, Aishwarya; Gan, Zhe; Zhang, Pengchuan; Wang, Jianfeng; Li, Linjie; Liu, Zicheng; Liu, Ce; LeCun, Yann; Peng, Nanyun; et al (October 2022, NeurIPS)

   Vision-language (VL) pre-training has recently received considerable attention. However, most existing end-to-end pre-training approaches either only aim to tackle VL tasks such as image-text retrieval, visual question answering (VQA) and image captioning that test high-level understanding of images, or only target region-level understanding for tasks such as phrase grounding and object detection. We present FIBER (Fusion-In-the-Backbone-based transformER), a new VL model architecture that can seamlessly handle both these types of tasks. Instead of having dedicated transformer layers for fusion after the uni-modal backbones, FIBER pushes multimodal fusion deep into the model by inserting cross-attention into the image and text backbones to better capture multimodal interactions. In addition, unlike previous work that is either only pre-trained on image-text data or on fine-grained data with box-level annotations, we present a two-stage pre-training strategy that uses both these kinds of data efficiently: (i) coarse-grained pre-training based on image-text data; followed by (ii) fine-grained pre-training based on image-text-box data. We conduct comprehensive experiments on a wide range of VL tasks, ranging from VQA, image captioning, and retrieval, to phrase grounding, referring expression comprehension, and object detection. Using deep multimodal fusion coupled with the two-stage pre-training, FIBER provides consistent performance improvements over strong baselines across all tasks, often outperforming methods using magnitudes more data. Code is released at https://github.com/microsoft/FIBER. 

   more » « less