Answering questions about the spatial properties of the environment poses challenges for existing language and vision foundation models due to a lack of understanding of the 3D world notably in terms of relationships between objects. To push the field forward, multiple 3D Q&A datasets were proposed which, overall, provide a variety of questions, but they individually focus on particular aspects of 3D reasoning or are limited in terms of data modalities. To address this, we present Space3D-Bench - a collection of 1000 general spatial questions and answers related to scenes of the Replica dataset which offers a variety of data modalities: point clouds, posed RGB-D images, navigation meshes and 3D object detections. To ensure that the questions cover a wide range of 3D objectives, we propose an indoor spatial questions taxonomy inspired by geographic information systems and use it to balance the dataset accordingly. Moreover, we provide an assessment system that grades natural language responses based on predefined ground-truth answers by leveraging a Vision Language Model's comprehension of both text and images to compare the responses with ground-truth textual information or relevant visual data. Finally, we introduce a baseline called RAG3D-Chat integrating the world understanding of foundation models with rich context retrieval, achieving an accuracy of 67% on the proposed dataset.
Multimodal Large Language Models (MLLMs) demonstrate a strong understanding of the real world and can even handle complex tasks. However, they still fail on some straightforward visual question-answering (VQA) problems. This paper dives deeper into this issue, revealing that models tend to err when answering easy questions (e.g. Yes/No questions) about an image, even though they can correctly describe it. We refer to this model behavior discrepancy between difficult and simple questions as model laziness. To systematically investigate model laziness, we manually construct LazyBench, a benchmark that includes Yes/No, multiple choice, short answer questions, and image description tasks that are related to the same subjects in the images. Based on LazyBench, we observe that laziness widely exists in current advanced MLLMs (e.g. GPT-4o, Gemini-1.5-pro, Claude 3 and LLaVA-v1.5-13B), and it is more pronounced on stronger models. We also analyze the VQA v2 (LLaVA-v1.5-13B) benchmark and find that about half of its failure cases are caused by model laziness, which further highlights the importance of ensuring that the model fully utilizes its capability. To this end, we conduct preliminary exploration on how to mitigate laziness and find that chain of thought (CoT) can effectively address this issue.
Visual Word Sense Disambiguation (VWSD) is a novel challenging task with the goal of retrieving an image among a set of candidates, which better represents the meaning of an ambiguous word within a given context. In this paper, we make a substantial step towards unveiling this interesting task by applying a varying set of approaches. Since VWSD is primarily a text-image retrieval task, we explore the latest transformer-based methods for multimodal retrieval. Additionally, we utilize Large Language Models (LLMs) as knowledge bases to enhance the given phrases and resolve ambiguity related to the target word. We also study VWSD as a unimodal problem by converting to text-to-text and image-to-image retrieval, as well as question-answering (QA), to fully explore the capabilities of relevant models. To tap into the implicit knowledge of LLMs, we experiment with Chain-of-Thought (CoT) prompting to guide explainable answer generation. On top of all, we train a learn to rank (LTR) model in order to combine our different modules, achieving competitive ranking results. Extensive experiments on VWSD demonstrate valuable insights to effectively drive future directions.
Vision-Language Foundation Models (VLFMs) have made remarkable progress on various multimodal tasks, such as image captioning, image-text retrieval, visual question answering, and visual grounding. However, most methods rely on training with general image datasets, and the lack of geospatial data leads to poor performance on earth observation. Numerous geospatial image-text pair datasets and VLFMs fine-tuned on them have been proposed recently. These new approaches aim to leverage large-scale, multimodal geospatial data to build versatile intelligent models with diverse geo-perceptive capabilities, which we refer to as Vision-Language Geo-Foundation Models (VLGFMs). This paper thoroughly reviews VLGFMs, summarizing and analyzing recent developments in the field. In particular, we introduce the background and motivation behind the rise of VLGFMs, highlighting their unique research significance. Then, we systematically summarize the core technologies employed in VLGFMs, including data construction, model architectures, and applications of various multimodal geospatial tasks. Finally, we conclude with insights, issues, and discussions regarding future research directions. To the best of our knowledge, this is the first comprehensive literature review of VLGFMs. We keep tracing related works at https://github.com/zytx121/Awesome-VLGFM.
State-of-the-art large multi-modal models (LMMs) face challenges when processing high-resolution images, as these inputs are converted into enormous visual tokens, many of which are irrelevant to the downstream task. In this paper, we propose Multi-turn Grounding-based Policy Optimization (MGPO), an end-to-end reinforcement learning (RL) framework that enables LMMs to iteratively focus on key visual regions by automatically cropping sub-images, based on model-predicted grounding coordinates within a multi-turn conversation framework. Compared to supervised fine-tuning (SFT), which requires costly additional grounding annotations, our approach highlights that LMMs can emerge robust grounding abilities during the RL training process, leveraging only a binary reward function derived from the correctness of the final answer. Additionally, we observe that LMMs struggle to autonomously trigger visual grounding during the rollout process. To address this cold start problem, we design a multi-turn conversational template and restrict policy loss computation to model outputs generated across multiple dialogue rounds, thereby promoting stable optimization. Extensive experiments demonstrate that, when trained on standard visual-question-short answering data without grounding annotations, MGPO effectively elicits stronger grounding capabilities compared to GRPO, leading to 5.4\% improvement on in-distribution MME-Realworld and 5.2\% improvement on the challenging out-of-distribution (OOD) V* Bench. Notably, MGPO post-training on Qwen2.5-VL-7B with 21K samples surpasses OpenAI's o1 and GPT-4o models on the OOD V* Bench. Codes are available at https://github.com/EvolvingLMMs-Lab/MGPO.
Multi-modal reasoning systems rely on a pre-trained object detector to extract regions of interest from the image. However, this crucial module is typically used as a black box, trained independently of the downstream task and on a fixed vocabulary of objects and attributes. This makes it challenging for such systems to capture the long tail of visual concepts expressed in free form text. In this paper we propose MDETR, an end-to-end modulated detector that detects objects in an image conditioned on a raw text query, like a caption or a question. We use a transformer-based architecture to reason jointly over text and image by fusing the two modalities at an early stage of the model. We pre-train the network on 1.3M text-image pairs, mined from pre-existing multi-modal datasets having explicit alignment between phrases in text and objects in the image. We then fine-tune on several downstream tasks such as phrase grounding, referring expression comprehension and segmentation, achieving state-of-the-art results on popular benchmarks. We also investigate the utility of our model as an object detector on a given label set when fine-tuned in a few-shot setting. We show that our pre-training approach provides a way to handle the long tail of object categories which have very few labelled instances. Our approach can be easily extended for visual question answering, achieving competitive performance on GQA and CLEVR. The code and models are available at https://github.com/ashkamath/mdetr.
There has been an explosion of work in the vision & language community during the past few years from image captioning to video transcription, and answering questions about images. These tasks have focused on literal descriptions of the image. To move beyond the literal, we choose to explore how questions about an image are often directed at commonsense inference and the abstract events evoked by objects in the image. In this paper, we introduce the novel task of Visual Question Generation (VQG), where the system is tasked with asking a natural and engaging question when shown an image. We provide three datasets which cover a variety of images from object-centric to event-centric, with considerably more abstract training data than provided to state-of-the-art captioning systems thus far. We train and test several generative and retrieval models to tackle the task of VQG. Evaluation results show that while such models ask reasonable questions for a variety of images, there is still a wide gap with human performance which motivates further work on connecting images with commonsense knowledge and pragmatics. Our proposed task offers a new challenge to the community which we hope furthers interest in exploring deeper connections between vision & language.
Joint image-text embedding extracted from medical images and associated contextual reports is the bedrock for most biomedical vision-and-language (V+L) tasks, including medical visual question answering, clinical image-text retrieval, clinical report auto-generation. In this study, we adopt four pre-trained V+L models: LXMERT, VisualBERT, UNIER and PixelBERT to learn multimodal representation from MIMIC-CXR radiographs and associated reports. The extrinsic evaluation on OpenI dataset shows that in comparison to the pioneering CNN-RNN model, the joint embedding learned by pre-trained V+L models demonstrate performance improvement in the thoracic findings classification task. We conduct an ablation study to analyze the contribution of certain model components and validate the advantage of joint embedding over text-only embedding. We also visualize attention maps to illustrate the attention mechanism of V+L models.
Summarization of multimedia data becomes increasingly significant as it is the basis for many real-world applications, such as question answering, Web search, and so forth. Most existing multi-modal summarization works however have used visual complementary features extracted from images rather than videos, thereby losing abundant information. Hence, we propose a novel multi-modal summarization task to summarize from a document and its associated video. In this work, we also build a baseline general model with effective strategies, i.e., bi-hop attention and improved late fusion mechanisms to bridge the gap between different modalities, and a bi-stream summarization strategy to employ text and video summarization simultaneously. Comprehensive experiments show that the proposed model is beneficial for multi-modal summarization and superior to existing methods. Moreover, we collect a novel dataset and it provides a new resource for future study that results from documents and videos.
We are interested in image manipulation via natural language text -- a task that is useful for multiple AI applications but requires complex reasoning over multi-modal spaces. We extend recently proposed Neuro Symbolic Concept Learning (NSCL), which has been quite effective for the task of Visual Question Answering (VQA), for the task of image manipulation. Our system referred to as NeuroSIM can perform complex multi-hop reasoning over multi-object scenes and only requires weak supervision in the form of annotated data for VQA. NeuroSIM parses an instruction into a symbolic program, based on a Domain Specific Language (DSL) comprising of object attributes and manipulation operations, that guides its execution. We create a new dataset for the task, and extensive experiments demonstrate that NeuroSIM is highly competitive with or beats SOTA baselines that make use of supervised data for manipulation.
Large language models (LLMs) have achieved state-of-the-art results in many natural language processing tasks. They have also demonstrated ability to adapt well to different tasks through zero-shot or few-shot settings. With the capability of these LLMs, researchers have looked into how to adopt them for use with Visual Question Answering (VQA). Many methods require further training to align the image and text embeddings. However, these methods are computationally expensive and requires large scale image-text dataset for training. In this paper, we explore a method of combining pretrained LLMs and other foundation models without further training to solve the VQA problem. The general idea is to use natural language to represent the images such that the LLM can understand the images. We explore different decoding strategies for generating textual representation of the image and evaluate their performance on the VQAv2 dataset.
Smaller-scale Vision-Langauge Models (VLMs) often claim to perform on par with larger models in general-domain visual grounding and question-answering benchmarks while offering advantages in computational efficiency and storage. However, their ability to handle rare objects, which fall into the long tail of data distributions, is less understood. To rigorously evaluate this aspect, we introduce the "Uncontextualized Uncommon Objects" (UOUO) benchmark. This benchmark focuses on systematically testing VLMs with both large and small parameter counts on rare and specialized objects. Our comprehensive analysis reveals that while smaller VLMs maintain competitive performance on common datasets, they significantly underperform on tasks involving uncommon objects. We also propose an advanced, scalable pipeline for data collection and cleaning, ensuring the UOUO benchmark provides high-quality, challenging instances. These findings highlight the need to consider long-tail distributions when assessing the true capabilities of VLMs.
Although recent text-to-image generative models have achieved impressive performance, they still often struggle with capturing the compositional complexities of prompts including attribute binding, and spatial relationships between different entities. This misalignment is not revealed by common evaluation metrics such as CLIPScore. Recent works have proposed evaluation metrics that utilize Visual Question Answering (VQA) by decomposing prompts into questions about the generated image for more robust compositional evaluation. Although these methods align better with human evaluations, they still fail to fully cover the compositionality within the image. To address this, we propose a novel metric that breaks down images into components, and texts into fine-grained questions about the generated image for evaluation. Our method outperforms previous state-of-the-art metrics, demonstrating its effectiveness in evaluating text-to-image generative models. Code is available at https://github.com/hadi-hosseini/ T2I-FineEval.
Previous research in human gesture recognition has largely overlooked multi-person interactions, which are crucial for understanding the social context of naturally occurring gestures. This limitation in existing datasets presents a significant challenge in aligning human gestures with other modalities like language and speech. To address this issue, we introduce SocialGesture, the first large-scale dataset specifically designed for multi-person gesture analysis. SocialGesture features a diverse range of natural scenarios and supports multiple gesture analysis tasks, including video-based recognition and temporal localization, providing a valuable resource for advancing the study of gesture during complex social interactions. Furthermore, we propose a novel visual question answering (VQA) task to benchmark vision language models'(VLMs) performance on social gesture understanding. Our findings highlight several limitations of current gesture recognition models, offering insights into future directions for improvement in this field. SocialGesture is available at huggingface.co/datasets/IrohXu/SocialGesture.
Classically, visual object tracking involves following a target object throughout a given video, and it provides us the motion trajectory of the object. However, for many practical applications, this output is often insufficient since additional semantic information is required to act on the video material. Example applications of this are surveillance and target-specific video summarization, where the target needs to be monitored with respect to certain predefined constraints, e.g., 'when standing near a yellow car'. This paper explores, tracking visual objects subjected to additional lingual constraints. Differently from Li et al., we impose additional lingual constraints upon tracking, which enables new applications of tracking. Whereas in their work the goal is to improve and extend upon tracking itself. To perform benchmarks and experiments, we contribute two datasets: c-MOT16 and c-LaSOT, curated through appending additional constraints to the frames of the original LaSOT and MOT16 datasets. We also experiment with two deep models SiamCT-DFG and SiamCT-CA, obtained through extending a recent state-of-the-art Siamese tracking method and adding modules inspired from the fields of natural language processing and visual question answering. Through experimental results, we show that the proposed model SiamCT-CA can significantly outperform its counterparts. Furthermore, our method enables the selective compression of videos, based on the validity of the constraint.
In-context learning (ICL) involves reasoning from given contextual examples. As more modalities comes, this procedure is becoming more challenging as the interleaved input modalities convolutes the understanding process. This is exemplified by the observation that multimodal models often struggle to effectively extrapolate from contextual examples to perform ICL. To address these challenges, we introduce MultiModal In-conteXt Tuning (M$^2$IXT), a lightweight module to enhance the ICL capabilities of multimodal unified models. The proposed M$^2$IXT module perceives an expandable context window to incorporate various labeled examples of multiple modalities (e.g., text, image, and coordinates). It can be prepended to various multimodal unified models (e.g., OFA, Unival, LLaVA) of different architectures and trained via a mixed-tasks strategy to enable rapid few-shot adaption on multiple tasks and datasets. When tuned on as little as 50K multimodal data, M$^2$IXT can boost the few-shot ICL performance significantly (e.g., 18\% relative increase for OFA), and obtained state-of-the-art results across an array of tasks including visual question answering, image captioning, visual grounding, and visual entailment, while being considerably small in terms of model parameters (e.g., $\sim$$20\times$ smaller than Flamingo or MMICL), highlighting the flexibility and effectiveness of M$^2$IXT as a multimodal in-context learner.
Given the ubiquity of charts as a data analysis, visualization, and decision-making tool across industries and sciences, there has been a growing interest in developing pre-trained foundation models as well as general purpose instruction-tuned models for chart understanding and reasoning. However, existing methods suffer crucial drawbacks across two critical axes affecting the performance of chart representation models: they are trained on data generated from underlying data tables of the charts, ignoring the visual trends and patterns in chart images, and use weakly aligned vision-language backbone models for domain-specific training, limiting their generalizability when encountering charts in the wild. We address these important drawbacks and introduce ChartGemma, a novel chart understanding and reasoning model developed over PaliGemma. Rather than relying on underlying data tables, ChartGemma is trained on instruction-tuning data generated directly from chart images, thus capturing both high-level trends and low-level visual information from a diverse set of charts. Our simple approach achieves state-of-the-art results across $5$ benchmarks spanning chart summarization, question answering, and fact-checking, and our elaborate qualitative studies on real-world charts show that ChartGemma generates more realistic and factually correct summaries compared to its contemporaries. We release the code, model checkpoints, dataset, and demos at https://github.com/vis-nlp/ChartGemma.
시각적 다이어그램, 그림, 표는 과학 논문의 핵심이며, 텍스트로 포착되지 않는 정보를 전달한다. 시각장애 또는 저시력(BLV) 과학자들은 전통적으로 논문의 그림에 접근하기 위해 정적 대체 텍스트에 의존해 왔지만, 인공지능(AI)의 부상으로 대화형 질의응답(QA)이 시각적 탐험을 위한 실행 가능한 패러다임이 되었음에도, 과학자들이 실제로 시각 QA를 어떻게 사용하는지 또는 그 접근성을 어떻게 개선할 수 있는지에 대해서는 알려진 바가 거의 없다. 본 연구에서는 서로 다른 STEM 분야의 BLV 과학자 5명과 시각 정상 과학자 5명을 인터뷰하여, ChatGPT와 Gemini라는 두 AI 도구를 사용해 멀티모달 과학 문서를 질의하는 방식을 조사한다. 우리의 발견은 과학자들이 멀티모달 콘텐츠를 검토하는 방식, 시각 자료와 상호작용하기 위한 기존 관행(접근성 우회 방법 포함), 그리고 멀티모달 질의에 대한 AI 생성 응답의 적합성에 대한 피드백을 규명한다. 또한 모호하거나 불완전한 이미지 설명뿐만 아니라 더 넓게는 잘못된 AI 출력이 BLV 및 시각 정상 과학자 모두 AI 워크플로를 포기하게 할 수 있음을 발견한다. 향후 연구를 지원하기 위해, 참가자들이 각자의 분야 논문에 대해 AI 도구와 상호작용한 115개의 질의 및 응답 데이터셋을 추가로 기여한다. 마지막으로 AI 기반 과학 QA 시스템에 대한 함의를 논의하며, 다양한 능력과 분야에 걸친 접근성 고려사항을 강조한다.
Vision-language models (VLMs) have made significant progress in recent visual-question-answering (VQA) benchmarks that evaluate complex visio-linguistic reasoning. However, are these models truly effective? In this work, we show that VLMs still struggle with natural images and questions that humans can easily answer, which we term natural adversarial samples. We also find it surprisingly easy to generate these VQA samples from natural image-text corpora using off-the-shelf models like CLIP and ChatGPT. We propose a semi-automated approach to collect a new benchmark, NaturalBench, for reliably evaluating VLMs with 10,000 human-verified VQA samples. Crucially, we adopt a $\textbf{vision-centric}$ design by pairing each question with two images that yield different answers, preventing blind solutions from answering without using the images. This makes NaturalBench more challenging than previous benchmarks that can be solved with commonsense priors. We evaluate 53 state-of-the-art VLMs on NaturalBench, showing that models like LLaVA-OneVision, Cambrian-1, Llama3.2-Vision, Molmo, Qwen2-VL, and even GPT-4o lag 50%-70% behind human performance (over 90%). We analyze why NaturalBench is hard from two angles: (1) Compositionality: Solving NaturalBench requires diverse visio-linguistic skills, including understanding attribute bindings, object relationships, and advanced reasoning like logic and counting. To this end, unlike prior work that uses a single tag per sample, we tag each NaturalBench sample with 1 to 8 skill tags for fine-grained evaluation. (2) Biases: NaturalBench exposes severe biases in VLMs, as models often choose the same answer regardless of the image. Lastly, we apply our benchmark curation method to diverse data sources, including long captions (over 100 words) and non-English languages like Chinese and Hindi, highlighting its potential for dynamic evaluations of VLMs.
Tables contain factual and quantitative data accompanied by various structures and contents that pose challenges for machine comprehension. Previous methods generally design task-specific architectures and objectives for individual tasks, resulting in modal isolation and intricate workflows. In this paper, we present a novel large vision-language model, TabPedia, equipped with a concept synergy mechanism. In this mechanism, all the involved diverse visual table understanding (VTU) tasks and multi-source visual embeddings are abstracted as concepts. This unified framework allows TabPedia to seamlessly integrate VTU tasks, such as table detection, table structure recognition, table querying, and table question answering, by leveraging the capabilities of large language models (LLMs). Moreover, the concept synergy mechanism enables table perception-related and comprehension-related tasks to work in harmony, as they can effectively leverage the needed clues from the corresponding source perception embeddings. Furthermore, to better evaluate the VTU task in real-world scenarios, we establish a new and comprehensive table VQA benchmark, ComTQA, featuring approximately 9,000 QA pairs. Extensive quantitative and qualitative experiments on both table perception and comprehension tasks, conducted across various public benchmarks, validate the effectiveness of our TabPedia. The superior performance further confirms the feasibility of using LLMs for understanding visual tables when all concepts work in synergy. The benchmark ComTQA has been open-sourced at https://huggingface.co/datasets/ByteDance/ComTQA. The source code and model also have been released athttps://github.com/zhaowc-ustc/TabPedia.