Advances in GPT-based large language models (LLMs) are revolutionizing natural language processing, exponentially increasing its use across various domains. Incorporating uni-directional attention, these autoregressive LLMs can generate long and coherent paragraphs. However, for visual question answering (VQA) tasks that require both vision and language processing, models with bi-directional attention or models employing fusion techniques are often employed to capture the context of multiple modalities all at once. As GPT does not natively process vision tokens, to exploit the advancements in GPT models for VQA in robotic surgery, we design an end-to-end trainable Language-Vision GPT (LV-GPT) model that expands the GPT2 model to include vision input (image). The proposed LV-GPT incorporates a feature extractor (vision tokenizer) and vision token embedding (token type and pose). Given the limitations of unidirectional attention in GPT models and their ability to generate coherent long paragraphs, we carefully sequence the word tokens before vision tokens, mimicking the human thought process of understanding the question to infer an answer from an image. Quantitatively, we prove that the LV-GPT model outperforms other state-of-the-art VQA models on two publically available surgical-VQA datasets (based on endoscopic vision challenge robotic scene segmentation 2018 and CholecTriplet2021) and on our newly annotated dataset (based on the holistic surgical scene dataset). We further annotate all three datasets to include question-type annotations to allow sub-type analysis. Furthermore, we extensively study and present the effects of token sequencing, token type and pose embedding for vision tokens in the LV-GPT model.
Causal video question answering (QA) has garnered increasing interest, yet existing datasets often lack depth in causal reasoning. To address this gap, we capitalize on the unique properties of cartoons and construct CausalChaos!, a novel, challenging causal Why-QA dataset built upon the iconic "Tom and Jerry" cartoon series. Cartoons use the principles of animation that allow animators to create expressive, unambiguous causal relationships between events to form a coherent storyline. Utilizing these properties, along with thought-provoking questions and multi-level answers (answer and detailed causal explanation), our questions involve causal chains that interconnect multiple dynamic interactions between characters and visual scenes. These factors demand models to solve more challenging, yet well-defined causal relationships. We also introduce hard incorrect answer mining, including a causally confusing version that is even more challenging. While models perform well, there is much room for improvement, especially, on open-ended answers. We identify more advanced/explicit causal relationship modeling & joint modeling of vision and language as the immediate areas for future efforts to focus upon. Along with the other complementary datasets, our new challenging dataset will pave the way for these developments in the field.
The recent advancements in auto-regressive multimodal large language models (MLLMs) have demonstrated promising progress for vision-language tasks. While there exists a variety of studies investigating the processing of linguistic information within large language models, little is currently known about the inner working mechanism of MLLMs and how linguistic and visual information interact within these models. In this study, we aim to fill this gap by examining the information flow between different modalities -- language and vision -- in MLLMs, focusing on visual question answering. Specifically, given an image-question pair as input, we investigate where in the model and how the visual and linguistic information are combined to generate the final prediction. Conducting experiments with a series of models from the LLaVA series, we find that there are two distinct stages in the process of integration of the two modalities. In the lower layers, the model first transfers the more general visual features of the whole image into the representations of (linguistic) question tokens. In the middle layers, it once again transfers visual information about specific objects relevant to the question to the respective token positions of the question. Finally, in the higher layers, the resulting multimodal representation is propagated to the last position of the input sequence for the final prediction. Overall, our findings provide a new and comprehensive perspective on the spatial and functional aspects of image and language processing in the MLLMs, thereby facilitating future research into multimodal information localization and editing. Our code and collected dataset are released here: https://github.com/FightingFighting/cross-modal-information-flow-in-MLLM.git.
We introduce CLEVR-Math, a multi-modal math word problems dataset consisting of simple math word problems involving addition/subtraction, represented partly by a textual description and partly by an image illustrating the scenario. The text describes actions performed on the scene that is depicted in the image. Since the question posed may not be about the scene in the image, but about the state of the scene before or after the actions are applied, the solver envision or imagine the state changes due to these actions. Solving these word problems requires a combination of language, visual and mathematical reasoning. We apply state-of-the-art neural and neuro-symbolic models for visual question answering on CLEVR-Math and empirically evaluate their performances. Our results show how neither method generalise to chains of operations. We discuss the limitations of the two in addressing the task of multi-modal word problem solving.
Compositional generalization is the ability of generalizing novel compositions from seen primitives, and has received much attention in vision-and-language (V\&L) recently. Due to the multi-modal nature of V\&L tasks, the primitives composing compositions source from different modalities, resulting in multi-sourced novel compositions. However, the generalization ability over multi-sourced novel compositions, \textit{i.e.}, multi-sourced compositional generalization (MSCG) remains unexplored. In this paper, we explore MSCG in the context of visual question answering (VQA), and propose a retrieval-augmented training framework to enhance the MSCG ability of VQA models by learning unified representations for primitives from different modalities. Specifically, semantically equivalent primitives are retrieved for each primitive in the training samples, and the retrieved features are aggregated with the original primitive to refine the model. This process helps the model learn consistent representations for the same semantic primitives across different modalities. To evaluate the MSCG ability of VQA models, we construct a new GQA-MSCG dataset based on the GQA dataset, in which samples include three types of novel compositions composed of primitives from different modalities. Experimental results demonstrate the effectiveness of the proposed framework. We release GQA-MSCG at https://github.com/NeverMoreLCH/MSCG.
Memes are a widely popular tool for web users to express their thoughts using visual metaphors. Understanding memes requires recognizing and interpreting visual metaphors with respect to the text inside or around the meme, often while employing background knowledge and reasoning abilities. We present the task of meme captioning and release a new dataset, MemeCap. Our dataset contains 6.3K memes along with the title of the post containing the meme, the meme captions, the literal image caption, and the visual metaphors. Despite the recent success of vision and language (VL) models on tasks such as image captioning and visual question answering, our extensive experiments using state-of-the-art VL models show that they still struggle with visual metaphors, and perform substantially worse than humans.
Visual Question Answering (VQA) has benefited from increasingly sophisticated models, but has not enjoyed the same level of engagement in terms of data creation. In this paper, we propose a method that automatically derives VQA examples at volume, by leveraging the abundance of existing image-caption annotations combined with neural models for textual question generation. We show that the resulting data is of high-quality. VQA models trained on our data improve state-of-the-art zero-shot accuracy by double digits and achieve a level of robustness that lacks in the same model trained on human-annotated VQA data.
Instructional videos provide detailed how-to guides for various tasks, with viewers often posing questions regarding the content. Addressing these questions is vital for comprehending the content, yet receiving immediate answers is difficult. While numerous computational models have been developed for Video Question Answering (Video QA) tasks, they are primarily trained on questions generated based on video content, aiming to produce answers from within the content. However, in real-world situations, users may pose questions that go beyond the video's informational boundaries, highlighting the necessity to determine if a video can provide the answer. Discerning whether a question can be answered by video content is challenging due to the multi-modal nature of videos, where visual and verbal information are intertwined. To bridge this gap, we present the YTCommentQA dataset, which contains naturally-generated questions from YouTube, categorized by their answerability and required modality to answer -- visual, script, or both. Experiments with answerability classification tasks demonstrate the complexity of YTCommentQA and emphasize the need to comprehend the combined role of visual and script information in video reasoning. The dataset is available at https://github.com/lgresearch/YTCommentQA.
Many datasets have been shown to contain incidental correlations created by idiosyncrasies in the data collection process. For example, sentence entailment datasets can have spurious word-class correlations if nearly all contradiction sentences contain the word "not", and image recognition datasets can have tell-tale object-background correlations if dogs are always indoors. In this paper, we propose a method that can automatically detect and ignore these kinds of dataset-specific patterns, which we call dataset biases. Our method trains a lower capacity model in an ensemble with a higher capacity model. During training, the lower capacity model learns to capture relatively shallow correlations, which we hypothesize are likely to reflect dataset bias. This frees the higher capacity model to focus on patterns that should generalize better. We ensure the models learn non-overlapping approaches by introducing a novel method to make them conditionally independent. Importantly, our approach does not require the bias to be known in advance. We evaluate performance on synthetic datasets, and four datasets built to penalize models that exploit known biases on textual entailment, visual question answering, and image recognition tasks. We show improvement in all settings, including a 10 point gain on the visual question answering dataset.
Large vision-language models (LVLMs) have recently achieved significant progress, demonstrating strong capabilities in open-world visual understanding. However, it is not yet clear how LVLMs address demographic biases in real life, especially the disparities across attributes such as gender, skin tone, age and race. In this paper, We empirically investigate visual fairness in several mainstream LVLMs by auditing their performance disparities across demographic attributes using public fairness benchmark datasets (e.g., FACET, UTKFace). Our fairness evaluation framework employs direct and single-choice question prompt on visual question-answering/classification tasks. Despite advancements in visual understanding, our zero-shot prompting results show that both open-source and closed-source LVLMs continue to exhibit fairness issues across different prompts and demographic groups. Furthermore, we propose a potential multi-modal Chain-of-thought (CoT) based strategy for bias mitigation, applicable to both open-source and closed-source LVLMs. This approach enhances transparency and offers a scalable solution for addressing fairness, providing a solid foundation for future bias reduction efforts.
대형 다중모달 모델(Large Multimodal Models, LMMs)은 시각-언어 및 시각 중심 과제에서 뛰어난 성능을 보이나, 복합적 구성 추론(compositional reasoning) 능력 통합에는 한계가 있습니다. 이를 개선하기 위해, 본 연구는 LMM이 내재된 시각 이해 및 그라운딩(grounding) 능력을 활용하여 복잡한 문제를 한 번의 순전파(forwarding)로 인간과 유사한 이해-사고-응답 과정을 수행하도록 하는 통합 시각 추론 메커니즘을 제안합니다. 또한, 33만 4천 개의 시각 지시문 데이터를 구축하여 다양한 장면과 다중 기초 시각 능력을 포함한 학습을 진행하였으며, 이를 통해 개발된 Griffon-R 모델은 복합 시각 추론 벤치마크에서 우수한 성능을 보이고 다중모달 능력 전반을 향상시켰습니다. 데이터와 모델, 코드는 곧 공개될 예정입니다.
Visual question answering (VQA) has the potential to make the Internet more accessible in an interactive way, allowing people who cannot see images to ask questions about them. However, multiple studies have shown that people who are blind or have low-vision prefer image explanations that incorporate the context in which an image appears, yet current VQA datasets focus on images in isolation. We argue that VQA models will not fully succeed at meeting people's needs unless they take context into account. To further motivate and analyze the distinction between different contexts, we introduce Context-VQA, a VQA dataset that pairs images with contexts, specifically types of websites (e.g., a shopping website). We find that the types of questions vary systematically across contexts. For example, images presented in a travel context garner 2 times more "Where?" questions, and images on social media and news garner 2.8 and 1.8 times more "Who?" questions than the average. We also find that context effects are especially important when participants can't see the image. These results demonstrate that context affects the types of questions asked and that VQA models should be context-sensitive to better meet people's needs, especially in accessibility settings.
Seeking answers effectively for long videos is essential to build video question answering (videoQA) systems. Previous methods adaptively select frames and regions from long videos to save computations. However, this fails to reason over the whole sequence of video, leading to sub-optimal performance. To address this problem, we introduce a state space layer (SSL) into multi-modal Transformer to efficiently integrate global semantics of the video, which mitigates the video information loss caused by frame and region selection modules. Our SSL includes a gating unit to enable controllability over the flow of global semantics into visual representations. To further enhance the controllability, we introduce a cross-modal compositional congruence (C^3) objective to encourage global semantics aligned with the question. To rigorously evaluate long-form videoQA capacity, we construct two new benchmarks Ego-QA and MAD-QA featuring videos of considerably long length, i.e. 17.5 minutes and 1.9 hours, respectively. Extensive experiments demonstrate the superiority of our framework on these new as well as existing datasets. The code, model, and data have been made available at https://nguyentthong.github.io/Long_form_VideoQA.
Models for Visual Question Answering (VQA) often rely on the spurious correlations, i.e., the language priors, that appear in the biased samples of training set, which make them brittle against the out-of-distribution (OOD) test data. Recent methods have achieved promising progress in overcoming this problem by reducing the impact of biased samples on model training. However, these models reveal a trade-off that the improvements on OOD data severely sacrifice the performance on the in-distribution (ID) data (which is dominated by the biased samples). Therefore, we propose a novel contrastive learning approach, MMBS, for building robust VQA models by Making the Most of Biased Samples. Specifically, we construct positive samples for contrastive learning by eliminating the information related to spurious correlation from the original training samples and explore several strategies to use the constructed positive samples for training. Instead of undermining the importance of biased samples in model training, our approach precisely exploits the biased samples for unbiased information that contributes to reasoning. The proposed method is compatible with various VQA backbones. We validate our contributions by achieving competitive performance on the OOD dataset VQA-CP v2 while preserving robust performance on the ID dataset VQA v2.
Thanks to the emerging of foundation models, the large language and vision models are integrated to acquire the multimodal ability of visual captioning, question answering, etc. Although existing multimodal models present impressive performance of visual understanding and reasoning, their limits are still largely under-explored due to the scarcity of high-quality instruction tuning data. To push the limits of multimodal capability, we Scale up Visual Instruction Tuning (SVIT) by constructing a dataset of 4.2 million visual instruction tuning data including 1.6M conversation question-answer (QA) pairs, 1.6M complex reasoning QA pairs, 1.0M referring QA pairs and 106K detailed image descriptions. Besides the volume, the proposed dataset is also featured by the high quality and rich diversity, which is generated by prompting GPT-4 with the abundant manual annotations of images. We also propose a new data recipe to select subset with better diversity and balance, which evokes model's superior capabilities. Extensive experiments verify that SVIT-v1.5, trained on the proposed dataset, outperforms state-of-the-art Multimodal Large Language Models on popular benchmarks. The data and code are publicly available at https://github.com/BAAI-DCAI/Visual-Instruction-Tuning.
Recently, there has been an increasing interest in building question answering (QA) models that reason across multiple modalities, such as text and images. However, QA using images is often limited to just picking the answer from a pre-defined set of options. In addition, images in the real world, especially in news, have objects that are co-referential to the text, with complementary information from both modalities. In this paper, we present a new QA evaluation benchmark with 1,384 questions over news articles that require cross-media grounding of objects in images onto text. Specifically, the task involves multi-hop questions that require reasoning over image-caption pairs to identify the grounded visual object being referred to and then predicting a span from the news body text to answer the question. In addition, we introduce a novel multimedia data augmentation framework, based on cross-media knowledge extraction and synthetic question-answer generation, to automatically augment data that can provide weak supervision for this task. We evaluate both pipeline-based and end-to-end pretraining-based multimedia QA models on our benchmark, and show that they achieve promising performance, while considerably lagging behind human performance hence leaving large room for future work on this challenging new task.
The success of deep learning has enabled advances in multimodal tasks that require non-trivial fusion of multiple input domains. Although multimodal models have shown potential in many problems, their increased complexity makes them more vulnerable to attacks. A Backdoor (or Trojan) attack is a class of security vulnerability wherein an attacker embeds a malicious secret behavior into a network (e.g. targeted misclassification) that is activated when an attacker-specified trigger is added to an input. In this work, we show that multimodal networks are vulnerable to a novel type of attack that we refer to as Dual-Key Multimodal Backdoors. This attack exploits the complex fusion mechanisms used by state-of-the-art networks to embed backdoors that are both effective and stealthy. Instead of using a single trigger, the proposed attack embeds a trigger in each of the input modalities and activates the malicious behavior only when both the triggers are present. We present an extensive study of multimodal backdoors on the Visual Question Answering (VQA) task with multiple architectures and visual feature backbones. A major challenge in embedding backdoors in VQA models is that most models use visual features extracted from a fixed pretrained object detector. This is challenging for the attacker as the detector can distort or ignore the visual trigger entirely, which leads to models where backdoors are over-reliant on the language trigger. We tackle this problem by proposing a visual trigger optimization strategy designed for pretrained object detectors. Through this method, we create Dual-Key Backdoors with over a 98% attack success rate while only poisoning 1% of the training data. Finally, we release TrojVQA, a large collection of clean and trojan VQA models to enable research in defending against multimodal backdoors.
This paper introduces a novel and significant challenge for Vision Language Models (VLMs), termed Unsolvable Problem Detection (UPD). UPD examines the VLM's ability to withhold answers when faced with unsolvable problems in the context of Visual Question Answering (VQA) tasks. UPD encompasses three distinct settings: Absent Answer Detection (AAD), Incompatible Answer Set Detection (IASD), and Incompatible Visual Question Detection (IVQD). To deeply investigate the UPD problem, extensive experiments indicate that most VLMs, including GPT-4V and LLaVA-Next-34B, struggle with our benchmarks to varying extents, highlighting significant room for the improvements. To address UPD, we explore both training-free and training-based solutions, offering new insights into their effectiveness and limitations. We hope our insights, together with future efforts within the proposed UPD settings, will enhance the broader understanding and development of more practical and reliable VLMs.
Knowledge-based visual question answering (VQA) involves answering questions that require external knowledge not present in the image. Existing methods first retrieve knowledge from external resources, then reason over the selected knowledge, the input image, and question for answer prediction. However, this two-step approach could lead to mismatches that potentially limit the VQA performance. For example, the retrieved knowledge might be noisy and irrelevant to the question, and the re-embedded knowledge features during reasoning might deviate from their original meanings in the knowledge base (KB). To address this challenge, we propose PICa, a simple yet effective method that Prompts GPT3 via the use of Image Captions, for knowledge-based VQA. Inspired by GPT-3's power in knowledge retrieval and question answering, instead of using structured KBs as in previous work, we treat GPT-3 as an implicit and unstructured KB that can jointly acquire and process relevant knowledge. Specifically, we first convert the image into captions (or tags) that GPT-3 can understand, then adapt GPT-3 to solve the VQA task in a few-shot manner by just providing a few in-context VQA examples. We further boost performance by carefully investigating: (i) what text formats best describe the image content, and (ii) how in-context examples can be better selected and used. PICa unlocks the first use of GPT-3 for multimodal tasks. By using only 16 examples, PICa surpasses the supervised state of the art by an absolute +8.6 points on the OK-VQA dataset. We also benchmark PICa on VQAv2, where PICa also shows a decent few-shot performance.
Vision-and-language reasoning requires an understanding of visual concepts, language semantics, and, most importantly, the alignment and relationships between these two modalities. We thus propose the LXMERT (Learning Cross-Modality Encoder Representations from Transformers) framework to learn these vision-and-language connections. In LXMERT, we build a large-scale Transformer model that consists of three encoders: an object relationship encoder, a language encoder, and a cross-modality encoder. Next, to endow our model with the capability of connecting vision and language semantics, we pre-train the model with large amounts of image-and-sentence pairs, via five diverse representative pre-training tasks: masked language modeling, masked object prediction (feature regression and label classification), cross-modality matching, and image question answering. These tasks help in learning both intra-modality and cross-modality relationships. After fine-tuning from our pre-trained parameters, our model achieves the state-of-the-art results on two visual question answering datasets (i.e., VQA and GQA). We also show the generalizability of our pre-trained cross-modality model by adapting it to a challenging visual-reasoning task, NLVR2, and improve the previous best result by 22% absolute (54% to 76%). Lastly, we demonstrate detailed ablation studies to prove that both our novel model components and pre-training strategies significantly contribute to our strong results; and also present several attention visualizations for the different encoders. Code and pre-trained models publicly available at: https://github.com/airsplay/lxmert