Search

Multimodal Emotion Recognition (MER) is an important research topic. This paper advocates for a transformative paradigm in MER. The rationale behind our work is that current approaches often rely on a limited set of basic emotion labels, which do not adequately represent the rich spectrum of human emotions. These traditional and overly simplistic emotion categories fail to capture the inherent complexity and subtlety of human emotional experiences, leading to limited generalizability and practicality. Therefore, we propose a new MER paradigm called Open-vocabulary MER (OV-MER), which encompasses a broader range of emotion labels to reflect the richness of human emotions. This paradigm relaxes the label space, allowing for the prediction of arbitrary numbers and categories of emotions. To support this transition, we provide a comprehensive solution that includes a newly constructed database based on LLM and human collaborative annotations, along with corresponding metrics and a series of benchmarks. We hope this work advances emotion recognition from basic emotions to more nuanced emotions, contributing to the development of emotional AI.

In the era of large language models (LLMs), the task of ``System I''~-~the fast, unconscious, and intuitive tasks, e.g., sentiment analysis, text classification, etc., have been argued to be successfully solved. However, sarcasm, as a subtle linguistic phenomenon, often employs rhetorical devices like hyperbole and figuration to convey true sentiments and intentions, involving a higher level of abstraction than sentiment analysis. There is growing concern that the argument about LLMs' success may not be fully tenable when considering sarcasm understanding. To address this question, we select eleven SOTA LLMs and eight SOTA pre-trained language models (PLMs) and present comprehensive evaluations on six widely used benchmark datasets through different prompting approaches, i.e., zero-shot input/output (IO) prompting, few-shot IO prompting, chain of thought (CoT) prompting. Our results highlight three key findings: (1) current LLMs underperform supervised PLMs based sarcasm detection baselines across six sarcasm benchmarks. This suggests that significant efforts are still required to improve LLMs' understanding of human sarcasm. (2) GPT-4 consistently and significantly outperforms other LLMs across various prompting methods, with an average improvement of 14.0\%$\uparrow$. Claude 3 and ChatGPT demonstrate the next best performance after GPT-4. (3) Few-shot IO prompting method outperforms the other two methods: zero-shot IO and few-shot CoT. The reason is that sarcasm detection, being a holistic, intuitive, and non-rational cognitive process, is argued not to adhere to step-by-step logical reasoning, making CoT less effective in understanding sarcasm compared to its effectiveness in mathematical reasoning tasks.

Explainable Multimodal Emotion Recognition (EMER) is an emerging task that aims to achieve reliable and accurate emotion recognition. However, due to the high annotation cost, the existing dataset (denoted as EMER-Fine) is small, making it difficult to perform supervised training. To reduce the annotation cost and expand the dataset size, this paper reviews the previous dataset construction process. Then, we simplify the annotation pipeline, avoid manual checks, and replace the closed-source models with open-source models. Finally, we build \textbf{EMER-Coarse}, a coarsely-labeled dataset containing large-scale samples. Besides the dataset, we propose a two-stage training framework \textbf{AffectGPT}. The first stage exploits EMER-Coarse to learn a coarse mapping between multimodal inputs and emotion-related descriptions; the second stage uses EMER-Fine to better align with manually-checked results. Experimental results demonstrate the effectiveness of our proposed method on the challenging EMER task. To facilitate further research, we will make the code and dataset available at: https://github.com/zeroQiaoba/AffectGPT.

Emotion and Intent Joint Understanding in Multimodal Conversation (MC-EIU) aims to decode the semantic information manifested in a multimodal conversational history, while inferring the emotions and intents simultaneously for the current utterance. MC-EIU is enabling technology for many human-computer interfaces. However, there is a lack of available datasets in terms of annotation, modality, language diversity, and accessibility. In this work, we propose an MC-EIU dataset, which features 7 emotion categories, 9 intent categories, 3 modalities, i.e., textual, acoustic, and visual content, and two languages, i.e., English and Mandarin. Furthermore, it is completely open-source for free access. To our knowledge, MC-EIU is the first comprehensive and rich emotion and intent joint understanding dataset for multimodal conversation. Together with the release of the dataset, we also develop an Emotion and Intent Interaction (EI$^2$) network as a reference system by modeling the deep correlation between emotion and intent in the multimodal conversation. With comparative experiments and ablation studies, we demonstrate the effectiveness of the proposed EI$^2$ method on the MC-EIU dataset. The dataset and codes will be made available at: https://github.com/MC-EIU/MC-EIU., Comment: 26 pages, 8 figures, 12 tables, NeurIPS 2024 Dataset and Benchmark Track

Accurate emotion perception is crucial for various applications, including human-computer interaction, education, and counseling. However, traditional single-modality approaches often fail to capture the complexity of real-world emotional expressions, which are inherently multimodal. Moreover, existing Multimodal Large Language Models (MLLMs) face challenges in integrating audio and recognizing subtle facial micro-expressions. To address this, we introduce the MERR dataset, containing 28,618 coarse-grained and 4,487 fine-grained annotated samples across diverse emotional categories. This dataset enables models to learn from varied scenarios and generalize to real-world applications. Furthermore, we propose Emotion-LLaMA, a model that seamlessly integrates audio, visual, and textual inputs through emotion-specific encoders. By aligning features into a shared space and employing a modified LLaMA model with instruction tuning, Emotion-LLaMA significantly enhances both emotional recognition and reasoning capabilities. Extensive evaluations show Emotion-LLaMA outperforms other MLLMs, achieving top scores in Clue Overlap (7.83) and Label Overlap (6.25) on EMER, an F1 score of 0.9036 on MER2023 challenge, and the highest UAR (45.59) and WAR (59.37) in zero-shot evaluations on DFEW dataset., Comment: 37 pages, 12 figures, Project: https://github.com/ZebangCheng/Emotion-LLaMA, Demo: https://huggingface.co/spaces/ZebangCheng/Emotion-LLaMA

Multimodal emotion recognition is an important research topic in artificial intelligence. Over the past few decades, researchers have made remarkable progress by increasing the dataset size and building more effective algorithms. However, due to problems such as complex environments and inaccurate annotations, current systems are hard to meet the demands of practical applications. Therefore, we organize the MER series of competitions to promote the development of this field. Last year, we launched MER2023, focusing on three interesting topics: multi-label learning, noise robustness, and semi-supervised learning. In this year's MER2024, besides expanding the dataset size, we further introduce a new track around open-vocabulary emotion recognition. The main purpose of this track is that existing datasets usually fix the label space and use majority voting to enhance the annotator consistency. However, this process may lead to inaccurate annotations, such as ignoring non-majority or non-candidate labels. In this track, we encourage participants to generate any number of labels in any category, aiming to describe emotional states as accurately as possible. Our baseline code relies on MERTools and is available at: https://github.com/zeroQiaoba/MERTools/tree/master/MER2024.

Multimodal fusion is a significant method for most multimodal tasks. With the recent surge in the number of large pre-trained models, combining both multimodal fusion methods and pre-trained model features can achieve outstanding performance in many multimodal tasks. In this paper, we present our approach, which leverages both advantages for addressing the task of Expression (Expr) Recognition and Valence-Arousal (VA) Estimation. We evaluate the Aff-Wild2 database using pre-trained models, then extract the final hidden layers of the models as features. Following preprocessing and interpolation or convolution to align the extracted features, different models are employed for modal fusion. Our code is available at GitHub - FulgenceWen/ABAW6th.

Deception detection has attracted increasing attention due to its importance in real-world scenarios. Its main goal is to detect deceptive behaviors from multimodal clues such as gestures, facial expressions, prosody, etc. However, these bases are usually subjective and related to personal habits. Therefore, we extend deception detection to deception reasoning, further providing objective evidence to support subjective judgment. Specifically, we provide potential lies and basic facts and then analyze why this sentence may be a lie by combining factual inconsistencies and intent behind them. Compared with deception detection, this task is more applicable to real-world scenarios. For example, in interrogation, the police should judge whether a person is lying based on solid evidence. This paper presents our initial attempts at this task, including constructing a dataset and defining evaluation metrics. Meanwhile, this task can serve as a benchmark for evaluating the complex reasoning capability of large language models. Our code and data are provided in the supplementary material.

Audio-Visual Emotion Recognition (AVER) has garnered increasing attention in recent years for its critical role in creating emotion-ware intelligent machines. Previous efforts in this area are dominated by the supervised learning paradigm. Despite significant progress, supervised learning is meeting its bottleneck due to the longstanding data scarcity issue in AVER. Motivated by recent advances in self-supervised learning, we propose Hierarchical Contrastive Masked Autoencoder (HiCMAE), a novel self-supervised framework that leverages large-scale self-supervised pre-training on vast unlabeled audio-visual data to promote the advancement of AVER. Following prior arts in self-supervised audio-visual representation learning, HiCMAE adopts two primary forms of self-supervision for pre-training, namely masked data modeling and contrastive learning. Unlike them which focus exclusively on top-layer representations while neglecting explicit guidance of intermediate layers, HiCMAE develops a three-pronged strategy to foster hierarchical audio-visual feature learning and improve the overall quality of learned representations. To verify the effectiveness of HiCMAE, we conduct extensive experiments on 9 datasets covering both categorical and dimensional AVER tasks. Experimental results show that our method significantly outperforms state-of-the-art supervised and self-supervised audio-visual methods, which indicates that HiCMAE is a powerful audio-visual emotion representation learner. Codes and models will be publicly available at https://github.com/sunlicai/HiCMAE., Comment: Accepted by Information Fusion. The code is available at https://github.com/sunlicai/HiCMAE

Multimodal emotion recognition plays a crucial role in enhancing user experience in human-computer interaction. Over the past few decades, researchers have proposed a series of algorithms and achieved impressive progress. Although each method shows its superior performance, different methods lack a fair comparison due to inconsistencies in feature extractors, evaluation manners, and experimental settings. These inconsistencies severely hinder the development of this field. Therefore, we build MERBench, a unified evaluation benchmark for multimodal emotion recognition. We aim to reveal the contribution of some important techniques employed in previous works, such as feature selection, multimodal fusion, robustness analysis, fine-tuning, pre-training, etc. We hope this benchmark can provide clear and comprehensive guidance for follow-up researchers. Based on the evaluation results of MERBench, we further point out some promising research directions. Additionally, we introduce a new emotion dataset MER2023, focusing on the Chinese language environment. This dataset can serve as a benchmark dataset for research on multi-label learning, noise robustness, and semi-supervised learning. We encourage the follow-up researchers to evaluate their algorithms under the same experimental setup as MERBench for fair comparisons. Our code is available at: https://github.com/zeroQiaoba/MERTools.

Video-based facial affect analysis has recently attracted increasing attention owing to its critical role in human-computer interaction. Previous studies mainly focus on developing various deep learning architectures and training them in a fully supervised manner. Although significant progress has been achieved by these supervised methods, the longstanding lack of large-scale high-quality labeled data severely hinders their further improvements. Motivated by the recent success of self-supervised learning in computer vision, this paper introduces a self-supervised approach, termed Self-supervised Video Facial Affect Perceiver (SVFAP), to address the dilemma faced by supervised methods. Specifically, SVFAP leverages masked facial video autoencoding to perform self-supervised pre-training on massive unlabeled facial videos. Considering that large spatiotemporal redundancy exists in facial videos, we propose a novel temporal pyramid and spatial bottleneck Transformer as the encoder of SVFAP, which not only largely reduces computational costs but also achieves excellent performance. To verify the effectiveness of our method, we conduct experiments on nine datasets spanning three downstream tasks, including dynamic facial expression recognition, dimensional emotion recognition, and personality recognition. Comprehensive results demonstrate that SVFAP can learn powerful affect-related representations via large-scale self-supervised pre-training and it significantly outperforms previous state-of-the-art methods on all datasets. Code is available at https://github.com/sunlicai/SVFAP., Comment: Published in: IEEE Transactions on Affective Computing (Early Access). The code and models are available at https://github.com/sunlicai/SVFAP

There remain two critical challenges that hinder the development of ERC. Firstly, there is a lack of exploration into mining deeper insights from the data itself for conversational emotion tasks. Secondly, the systems exhibit vulnerability to random modality feature missing, which is a common occurrence in realistic settings. Focusing on these two key challenges, we propose a novel framework for incomplete multimodal learning in ERC, called "Inverted Teacher-studEnt seArCH Network (ITEACH-Net)." ITEACH-Net comprises two novel components: the Emotion Context Changing Encoder (ECCE) and the Inverted Teacher-Student (ITS) framework. Specifically, leveraging the tendency for emotional states to exhibit local stability within conversational contexts, ECCE captures these patterns and further perceives their evolution over time. Recognizing the varying challenges of handling incomplete versus complete data, ITS employs a teacher-student framework to decouple the respective computations. Subsequently, through Neural Architecture Search, the student model develops enhanced computational capabilities for handling incomplete data compared to the teacher model. During testing, we design a novel evaluation method, testing the model's performance under different missing rate conditions without altering the model weights. We conduct experiments on three benchmark ERC datasets, and the results demonstrate that our ITEACH-Net outperforms existing methods in incomplete multimodal ERC. We believe ITEACH-Net can inspire relevant research on the intrinsic nature of emotions within conversation scenarios and pave a more robust route for incomplete learning techniques. Codes will be made available.

Recently, GPT-4 with Vision (GPT-4V) has demonstrated remarkable visual capabilities across various tasks, but its performance in emotion recognition has not been fully evaluated. To bridge this gap, we present the quantitative evaluation results of GPT-4V on 21 benchmark datasets covering 6 tasks: visual sentiment analysis, tweet sentiment analysis, micro-expression recognition, facial emotion recognition, dynamic facial emotion recognition, and multimodal emotion recognition. This paper collectively refers to these tasks as ``Generalized Emotion Recognition (GER)''. Through experimental analysis, we observe that GPT-4V exhibits strong visual understanding capabilities in GER tasks. Meanwhile, GPT-4V shows the ability to integrate multimodal clues and exploit temporal information, which is also critical for emotion recognition. However, it's worth noting that GPT-4V is primarily designed for general domains and cannot recognize micro-expressions that require specialized knowledge. To the best of our knowledge, this paper provides the first quantitative assessment of GPT-4V for GER tasks. We have open-sourced the code and encourage subsequent researchers to broaden the evaluation scope by including more tasks and datasets. Our code and evaluation results are available at: https://github.com/zeroQiaoba/gpt4v-emotion.

Multimodal emotion recognition (MER) in practical scenarios is significantly challenged by the presence of missing or incomplete data across different modalities. To overcome these challenges, researchers have aimed to simulate incomplete conditions during the training phase to enhance the system's overall robustness. Traditional methods have often involved discarding data or substituting data segments with zero vectors to approximate these incompletenesses. However, such approaches neither accurately represent real-world conditions nor adequately address the issue of noisy data availability. For instance, a blurry image cannot be simply replaced with zero vectors, while still retaining information. To tackle this issue and develop a more precise MER system, we introduce a novel noise-robust MER model that effectively learns robust multimodal joint representations from noisy data. This approach includes two pivotal components: firstly, a noise scheduler that adjusts the type and level of noise in the data to emulate various realistic incomplete situations. Secondly, a Variational AutoEncoder (VAE)-based module is employed to reconstruct these robust multimodal joint representations from the noisy inputs. Notably, the introduction of the noise scheduler enables the exploration of an entirely new type of incomplete data condition, which is impossible with existing methods. Extensive experimental evaluations on the benchmark datasets IEMOCAP and CMU-MOSEI demonstrate the effectiveness of the noise scheduler and the excellent performance of our proposed model. Our project is publicly available on https://github.com/WooyoohL/Noise-robust_MER., Comment: Accepted by MRAC@ACM Multimedia 2024

Dynamic facial expression recognition (DFER) is essential to the development of intelligent and empathetic machines. Prior efforts in this field mainly fall into supervised learning paradigm, which is severely restricted by the limited labeled data in existing datasets. Inspired by recent unprecedented success of masked autoencoders (e.g., VideoMAE), this paper proposes MAE-DFER, a novel self-supervised method which leverages large-scale self-supervised pre-training on abundant unlabeled data to largely advance the development of DFER. Since the vanilla Vision Transformer (ViT) employed in VideoMAE requires substantial computation during fine-tuning, MAE-DFER develops an efficient local-global interaction Transformer (LGI-Former) as the encoder. Moreover, in addition to the standalone appearance content reconstruction in VideoMAE, MAE-DFER also introduces explicit temporal facial motion modeling to encourage LGI-Former to excavate both static appearance and dynamic motion information. Extensive experiments on six datasets show that MAE-DFER consistently outperforms state-of-the-art supervised methods by significant margins (e.g., +6.30\% UAR on DFEW and +8.34\% UAR on MAFW), verifying that it can learn powerful dynamic facial representations via large-scale self-supervised pre-training. Besides, it has comparable or even better performance than VideoMAE, while largely reducing the computational cost (about 38\% FLOPs). We believe MAE-DFER has paved a new way for the advancement of DFER and can inspire more relevant research in this field and even other related tasks. Codes and models are publicly available at https://github.com/sunlicai/MAE-DFER., Comment: ACM MM 2023 (camera ready). Codes and models are publicly available at https://github.com/sunlicai/MAE-DFER

Multimodal emotion recognition is an important research topic in artificial intelligence, whose main goal is to integrate multimodal clues to identify human emotional states. Current works generally assume accurate labels for benchmark datasets and focus on developing more effective architectures. However, emotion annotation relies on subjective judgment. To obtain more reliable labels, existing datasets usually restrict the label space to some basic categories, then hire plenty of annotators and use majority voting to select the most likely label. However, this process may result in some correct but non-candidate or non-majority labels being ignored. To ensure reliability without ignoring subtle emotions, we propose a new task called ``Explainable Multimodal Emotion Recognition (EMER)''. Unlike traditional emotion recognition, EMER takes a step further by providing explanations for these predictions. Through this task, we can extract relatively reliable labels since each label has a certain basis. Meanwhile, we borrow large language models (LLMs) to disambiguate unimodal clues and generate more complete multimodal explanations. From them, we can extract richer emotions in an open-vocabulary manner. This paper presents our initial attempt at this task, including introducing a new dataset, establishing baselines, and defining evaluation metrics. In addition, EMER can serve as a benchmark task to evaluate the audio-video-text understanding performance of multimodal LLMs.

Speech Emotion Recognition (SER) is an important research topic in human-computer interaction. Many recent works focus on directly extracting emotional cues through pre-trained knowledge, frequently overlooking considerations of appropriateness and comprehensiveness. Therefore, we propose a novel framework for pre-training knowledge in SER, called Multi-perspective Fusion Search Network (MFSN). Considering comprehensiveness, we partition speech knowledge into Textual-related Emotional Content (TEC) and Speech-related Emotional Content (SEC), capturing cues from both semantic and acoustic perspectives, and we design a new architecture search space to fully leverage them. Considering appropriateness, we verify the efficacy of different modeling approaches in capturing SEC and fills the gap in current research. Experimental results on multiple datasets demonstrate the superiority of MFSN.

Audio deepfake detection is an emerging topic in the artificial intelligence community. The second Audio Deepfake Detection Challenge (ADD 2023) aims to spur researchers around the world to build new innovative technologies that can further accelerate and foster research on detecting and analyzing deepfake speech utterances. Different from previous challenges (e.g. ADD 2022), ADD 2023 focuses on surpassing the constraints of binary real/fake classification, and actually localizing the manipulated intervals in a partially fake speech as well as pinpointing the source responsible for generating any fake audio. Furthermore, ADD 2023 includes more rounds of evaluation for the fake audio game sub-challenge. The ADD 2023 challenge includes three subchallenges: audio fake game (FG), manipulation region location (RL) and deepfake algorithm recognition (AR). This paper describes the datasets, evaluation metrics, and protocols. Some findings are also reported in audio deepfake detection tasks.

The first Multimodal Emotion Recognition Challenge (MER 2023) was successfully held at ACM Multimedia. The challenge focuses on system robustness and consists of three distinct tracks: (1) MER-MULTI, where participants are required to recognize both discrete and dimensional emotions; (2) MER-NOISE, in which noise is added to test videos for modality robustness evaluation; (3) MER-SEMI, which provides a large amount of unlabeled samples for semi-supervised learning. In this paper, we introduce the motivation behind this challenge, describe the benchmark dataset, and provide some statistics about participants. To continue using this dataset after MER 2023, please sign a new End User License Agreement and send it to our official email address merchallenge.contact@gmail.com. We believe this high-quality dataset can become a new benchmark in multimodal emotion recognition, especially for the Chinese research community.

Partial-label learning (PLL) is an important branch of weakly supervised learning where the single ground truth resides in a set of candidate labels, while the research rarely considers the label imbalance. A recent study for imbalanced partial-Label learning proposed that the combinatorial challenge of partial-label learning and long-tail learning lies in matching between a decent marginal prior distribution with drawing the pseudo labels. However, we believe that even if the pseudo label matches the prior distribution, the tail classes will still be difficult to learn because the total weight is too small. Therefore, we propose a pseudo-label regularization technique specially designed for PLL. By punishing the pseudo labels of head classes, our method implements state-of-art under the standardized benchmarks compared to the previous PLL methods., Comment: arXiv admin note: substantial text overlap with arXiv:2209.10365 by other authors

Out-of-distribution (OOD) detection is critical to building reliable machine learning systems in the open world. Researchers have proposed various strategies to reduce model overconfidence on OOD data. Among them, ReAct is a typical and effective technique to deal with model overconfidence, which truncates high activations to increase the gap between in-distribution and OOD. Despite its promising results, is this technique the best choice for widening the gap? To answer this question, we leverage the variational method to find the optimal operation and verify the necessity of suppressing abnormally low and high activations and amplifying intermediate activations in OOD detection, rather than focusing only on high activations like ReAct. This motivates us to propose a novel technique called ``Variational Rectified Activation (VRA)'', which simulates these suppression and amplification operations using piecewise functions. Experimental results on multiple benchmark datasets demonstrate that our method outperforms existing post-hoc strategies. Meanwhile, VRA is compatible with different scoring functions and network architectures. \textcolor[rgb]{0.93,0.0,0.47}{Our code can be found in Supplementary Material}.

Noisy partial label learning (noisy PLL) is an important branch of weakly supervised learning. Unlike PLL where the ground-truth label must conceal in the candidate label set, noisy PLL relaxes this constraint and allows the ground-truth label may not be in the candidate label set. To address this challenging problem, most of the existing works attempt to detect noisy samples and estimate the ground-truth label for each noisy sample. However, detection errors are unavoidable. These errors can accumulate during training and continuously affect model optimization. To this end, we propose a novel framework for noisy PLL with theoretical guarantees, called ``Adjusting Label Importance Mechanism (ALIM)''. It aims to reduce the negative impact of detection errors by trading off the initial candidate set and model outputs. ALIM is a plug-in strategy that can be integrated with existing PLL approaches. Experimental results on benchmark datasets demonstrate that our method can achieve state-of-the-art performance on noisy PLL. \textcolor[rgb]{0.93,0.0,0.47}{Our code can be found in Supplementary Material}.

Partial label learning (PLL) is a typical weakly supervised learning, where each sample is associated with a set of candidate labels. The basic assumption of PLL is that the ground-truth label must reside in the candidate set. However, this assumption may not be satisfied due to the unprofessional judgment of the annotators, thus limiting the practical application of PLL. In this paper, we relax this assumption and focus on a more general problem, noisy PLL, where the ground-truth label may not exist in the candidate set. To address this challenging problem, we propose a novel framework called "Iterative Refinement Network (IRNet)". It aims to purify the noisy samples by two key modules, i.e., noisy sample detection and label correction. Ideally, we can convert noisy PLL into traditional PLL if all noisy samples are corrected. To guarantee the performance of these modules, we start with warm-up training and exploit data augmentation to reduce prediction errors. Through theoretical analysis, we prove that IRNet is able to reduce the noise level of the dataset and eventually approximate the Bayes optimal classifier. Experimental results on multiple benchmark datasets demonstrate the effectiveness of our method. IRNet is superior to existing state-of-the-art approaches on noisy PLL.

Medical Relation Extraction (MRE) task aims to extract relations between entities in medical texts. Traditional relation extraction methods achieve impressive success by exploring the syntactic information, e.g., dependency tree. However, the quality of the 1-best dependency tree for medical texts produced by an out-of-domain parser is relatively limited so that the performance of medical relation extraction method may degenerate. To this end, we propose a method to jointly model semantic and syntactic information from medical texts based on causal explanation theory. We generate dependency forests consisting of the semantic-embedded 1-best dependency tree. Then, a task-specific causal explainer is adopted to prune the dependency forests, which are further fed into a designed graph convolutional network to learn the corresponding representation for downstream task. Empirically, the various comparisons on benchmark medical datasets demonstrate the effectiveness of our model., Comment: Accepted to the conference of COLING2022 as an Oral presentation

With the proliferation of user-generated online videos, Multimodal Sentiment Analysis (MSA) has attracted increasing attention recently. Despite significant progress, there are still two major challenges on the way towards robust MSA: 1) inefficiency when modeling cross-modal interactions in unaligned multimodal data; and 2) vulnerability to random modality feature missing which typically occurs in realistic settings. In this paper, we propose a generic and unified framework to address them, named Efficient Multimodal Transformer with Dual-Level Feature Restoration (EMT-DLFR). Concretely, EMT employs utterance-level representations from each modality as the global multimodal context to interact with local unimodal features and mutually promote each other. It not only avoids the quadratic scaling cost of previous local-local cross-modal interaction methods but also leads to better performance. To improve model robustness in the incomplete modality setting, on the one hand, DLFR performs low-level feature reconstruction to implicitly encourage the model to learn semantic information from incomplete data. On the other hand, it innovatively regards complete and incomplete data as two different views of one sample and utilizes siamese representation learning to explicitly attract their high-level representations. Comprehensive experiments on three popular datasets demonstrate that our method achieves superior performance in both complete and incomplete modality settings., Comment: Accepted by TAC. The code is available at https://github.com/sunlicai/EMT-DLFR

In this paper, we propose the solution to the Multi-Task Learning (MTL) Challenge of the 4th Affective Behavior Analysis in-the-wild (ABAW) competition. The task of ABAW is to predict frame-level emotion descriptors from videos: discrete emotional state; valence and arousal; and action units. Although researchers have proposed several approaches and achieved promising results in ABAW, current works in this task rarely consider interactions between different emotion descriptors. To this end, we propose a novel end to end architecture to achieve full integration of different types of information. Experimental results demonstrate the effectiveness of our proposed solution.

Speech emotion recognition (SER) is an important research topic in human-computer interaction. Existing works mainly rely on human expertise to design models. Despite their success, different datasets often require distinct structures and hyperparameters. Searching for an optimal model for each dataset is time-consuming and labor-intensive. To address this problem, we propose a two-stream neural architecture search (NAS) based framework, called \enquote{EmotionNAS}. Specifically, we take two-stream features (i.e., handcrafted and deep features) as the inputs, followed by NAS to search for the optimal structure for each stream. Furthermore, we incorporate complementary information in different streams through an efficient information supplement module. Experimental results demonstrate that our method outperforms existing manually-designed and NAS-based models, setting the new state-of-the-art record., Comment: Accepted to Interspeech 2023

Conversations have become a critical data format on social media platforms. Understanding conversation from emotion, content and other aspects also attracts increasing attention from researchers due to its widespread application in human-computer interaction. In real-world environments, we often encounter the problem of incomplete modalities, which has become a core issue of conversation understanding. To address this problem, researchers propose various methods. However, existing approaches are mainly designed for individual utterances rather than conversational data, which cannot fully exploit temporal and speaker information in conversations. To this end, we propose a novel framework for incomplete multimodal learning in conversations, called "Graph Complete Network (GCNet)", filling the gap of existing works. Our GCNet contains two well-designed graph neural network-based modules, "Speaker GNN" and "Temporal GNN", to capture temporal and speaker dependencies. To make full use of complete and incomplete data, we jointly optimize classification and reconstruction tasks in an end-to-end manner. To verify the effectiveness of our method, we conduct experiments on three benchmark conversational datasets. Experimental results demonstrate that our GCNet is superior to existing state-of-the-art approaches in incomplete multimodal learning. Code is available at https://github.com/zeroQiaoba/GCNet.

Audio deepfake detection is an emerging topic, which was included in the ASVspoof 2021. However, the recent shared tasks have not covered many real-life and challenging scenarios. The first Audio Deep synthesis Detection challenge (ADD) was motivated to fill in the gap. The ADD 2022 includes three tracks: low-quality fake audio detection (LF), partially fake audio detection (PF) and audio fake game (FG). The LF track focuses on dealing with bona fide and fully fake utterances with various real-world noises etc. The PF track aims to distinguish the partially fake audio from the real. The FG track is a rivalry game, which includes two tasks: an audio generation task and an audio fake detection task. In this paper, we describe the datasets, evaluation metrics, and protocols. We also report major findings that reflect the recent advances in audio deepfake detection tasks., Comment: Accepted by ICASSP 2022

The conventional encoder-decoder framework for image captioning generally adopts a single-pass decoding process, which predicts the target descriptive sentence word by word in temporal order. Despite the great success of this framework, it still suffers from two serious disadvantages. Firstly, it is unable to correct the mistakes in the predicted words, which may mislead the subsequent prediction and result in error accumulation problem. Secondly, such a framework can only leverage the already generated words but not the possible future words, and thus lacks the ability of global planning on linguistic information. To overcome these limitations, we explore a universal two-pass decoding framework, where a single-pass decoding based model serving as the Drafting Model first generates a draft caption according to an input image, and a Deliberation Model then performs the polishing process to refine the draft caption to a better image description. Furthermore, inspired from the complementarity between different modalities, we propose a novel Cross Modification Attention (CMA) module to enhance the semantic expression of the image features and filter out error information from the draft captions. We integrate CMA with the decoder of our Deliberation Model and name it as Cross Modification Attention based Deliberation Model (CMA-DM). We train our proposed framework by jointly optimizing all trainable components from scratch with a trade-off coefficient. Experiments on MS COCO dataset demonstrate that our approach obtains significant improvements over single-pass decoding baselines and achieves competitive performances compared with other state-of-the-art two-pass decoding based methods., Comment: This work has been submitted to the IEEE TMM for possible publication

We introduce BEHAVIOR, a benchmark for embodied AI with 100 activities in simulation, spanning a range of everyday household chores such as cleaning, maintenance, and food preparation. These activities are designed to be realistic, diverse, and complex, aiming to reproduce the challenges that agents must face in the real world. Building such a benchmark poses three fundamental difficulties for each activity: definition (it can differ by time, place, or person), instantiation in a simulator, and evaluation. BEHAVIOR addresses these with three innovations. First, we propose an object-centric, predicate logic-based description language for expressing an activity's initial and goal conditions, enabling generation of diverse instances for any activity. Second, we identify the simulator-agnostic features required by an underlying environment to support BEHAVIOR, and demonstrate its realization in one such simulator. Third, we introduce a set of metrics to measure task progress and efficiency, absolute and relative to human demonstrators. We include 500 human demonstrations in virtual reality (VR) to serve as the human ground truth. Our experiments demonstrate that even state of the art embodied AI solutions struggle with the level of realism, diversity, and complexity imposed by the activities in our benchmark. We make BEHAVIOR publicly available at behavior.stanford.edu to facilitate and calibrate the development of new embodied AI solutions.