Even though there has been tremendous progress in the field of Visual Question Answering, models today still tend to be inconsistent and brittle. To this end, we propose a model-independent cyclic framework which increases consistency and robustness of any VQA architecture. We train our models to answer the original question, generate an implication based on the answer and then also learn to answer the generated implication correctly. As a part of the cyclic framework, we propose a novel implication generator which can generate implied questions from any question-answer pair. As a baseline for future works on consistency, we provide a new human annotated VQA-Implications dataset. The dataset consists of ~30k questions containing implications of 3 types - Logical Equivalence, Necessary Condition and Mutual Exclusion - made from the VQA v2.0 validation dataset. We show that our framework improves consistency of VQA models by ~15% on the rule-based dataset, ~7% on VQA-Implications dataset and robustness by ~2%, without degrading their performance. In addition, we also quantitatively show improvement in attention maps which highlights better multi-modal understanding of vision and language.
현재의 비디오 대규모 언어 모델(Video LLM)은 질의응답(QA)에서 뛰어난 성능을 보이지만, 대부분 시각적 근거 없이 텍스트 답변만 제공하는 블랙박스 방식으로 작동한다. 기존의 설명 가능성 연구들은 텍스트 기반 근거나 희소한 바운딩 박스에 의존하며, 폐색(occlusion)이나 비강체 변형(non-rigid deformation)과 같은 복잡한 비디오 동역학을 포착하는 데 한계가 있다. 우리는 근거 기반 비디오 질의응답(E-VQA)이라는 새로운 태스크를 제안한다. 이 태스크는 모델이 의미론적 답변과 함께 시공간적 근거—즉, 시간 구간과 조밀하게 추적된 객체 분할 마스크릿(masklet)—를 동시에 출력할 것을 요구한다. 이를 지원하기 위해 판별적(discriminative) 및 생성적(generative) 픽셀 수준 그라운딩을 위한 최초의 인간 검증 벤치마크인 ST-Evidence를 도입한다. 최신 모델들에 대한 평가 결과, QA 정확도와 실제 시각적 인식 사이의 심각한 괴리가 드러났으며, 이는 스케일링만으로는 해소되지 않는다. 이를 해결하기 위해 고수준 추론과 세밀한 그라운딩을 연결하는 16만 규모의 데이터셋인 ST-Evidence-Instruct를 생성하는 확장 가능한 자동화 생성 파이프라인을 개발한다. 이 데이터로 그라운딩 기반 Video LLM을 미세조정하면 동일한 크기의 UniPixel 기준선 대비 상당한 성능 향상을 달성한다(예: 7B 모델에서 t-mean +27.2, J&F +13.8). 이는 설명 가능하고 근거 기반의 비디오 이해를 위한 강건한 기준선을 확립한다. 코드와 데이터는 https://github.com/SalesforceAIResearch/EVQA 에서 공개되어 있다.
Deep neural networks have been critical in the task of Visual Question Answering (VQA), with research traditionally focused on improving model accuracy. Recently, however, there has been a trend towards evaluating the robustness of these models against adversarial attacks. This involves assessing the accuracy of VQA models under increasing levels of noise in the input, which can target either the image or the proposed query question, dubbed the main question. However, there is currently a lack of proper analysis of this aspect of VQA. This work proposes a new method that utilizes semantically related questions, referred to as basic questions, acting as noise to evaluate the robustness of VQA models. It is hypothesized that as the similarity of a basic question to the main question decreases, the level of noise increases. To generate a reasonable noise level for a given main question, a pool of basic questions is ranked based on their similarity to the main question, and this ranking problem is cast as a LASSO optimization problem. Additionally, this work proposes a novel robustness measure, R_score, and two basic question datasets to standardize the analysis of VQA model robustness. The experimental results demonstrate that the proposed evaluation method effectively analyzes the robustness of VQA models. Moreover, the experiments show that in-context learning with a chain of basic questions can enhance model accuracy.
The global shortage of healthcare workers has demanded the development of smart healthcare assistants, which can help monitor and alert healthcare workers when necessary. We examine the healthcare knowledge of existing Large Vision Language Models (LVLMs) via the Visual Question Answering (VQA) task in hospital settings through expert annotated open-ended questions. We introduce the Emergency Room Visual Question Answering (ERVQA) dataset, consisting of <image, question, answer> triplets covering diverse emergency room scenarios, a seminal benchmark for LVLMs. By developing a detailed error taxonomy and analyzing answer trends, we reveal the nuanced nature of the task. We benchmark state-of-the-art open-source and closed LVLMs using traditional and adapted VQA metrics: Entailment Score and CLIPScore Confidence. Analyzing errors across models, we infer trends based on properties like decoder type, model size, and in-context examples. Our findings suggest the ERVQA dataset presents a highly complex task, highlighting the need for specialized, domain-specific solutions.
Transformers for visual-language representation learning have been getting a lot of interest and shown tremendous performance on visual question answering (VQA) and grounding. But most systems that show good performance of those tasks still rely on pre-trained object detectors during training, which limits their applicability to the object classes available for those detectors. To mitigate this limitation, the following paper focuses on the problem of weakly supervised grounding in context of visual question answering in transformers. The approach leverages capsules by grouping each visual token in the visual encoder and uses activations from language self-attention layers as a text-guided selection module to mask those capsules before they are forwarded to the next layer. We evaluate our approach on the challenging GQA as well as VQA-HAT dataset for VQA grounding. Our experiments show that: while removing the information of masked objects from standard transformer architectures leads to a significant drop in performance, the integration of capsules significantly improves the grounding ability of such systems and provides new state-of-the-art results compared to other approaches in the field.
지식 기반 시각 질의응답(Knowledge-based Visual Question Answering)은 외부 지식을 시각 및 텍스트 정보와 통합하여 이미지에 관한 질문에 답하는 것을 목표로 한다. 최근 접근법들은 흔히 문맥 내 학습(in-context learning)에 의존하여, 제로샷 또는 퓨샷 방식으로 멀티모달 컨텍스트와 함께 대규모 언어 모델(LLM)에 프롬프트를 제공한다. 그러나 우리는 이질적인 시각 설명과 검색된 지식을 길고 구조화되지 않은 프롬프트로 단순히 이어 붙이면, 관련 없는 컨텍스트가 지나치게 많아지고 명시적인 관계 구조가 부족하여 추론 성능이 오히려 저하되는 경우가 많다는 점을 관찰하였다. 본 논문에서는 예측을 위해 명시적 관계와 암묵적 관계를 모두 추론하는 LLM 기반 구조화된 컨텍스트 추론(Structured Context Reasoning, SCoRe) 프레임워크를 제안한다. SCoRe는 세 단계로 구성된다. 첫째, 컨텍스트 획득(Context Acquisition) 단계에서는 다양한 시각 노트를 생성하고 효율적인 2단계 멀티모달 검색 전략을 통해 명시적 지식을 검색한다. 둘째, 컨텍스트 선택(Context Selection) 단계에서는 LLM이 안내하는 선택 방식을 이용해 관련 있는 시각 지식, 명시적 지식, 암묵적 지식을 걸러낸다. 셋째, 컨텍스트 압축(Context Compression) 단계에서는 관계 논리 증류(Relational Logic Distillation, RLD)를 수행하여 원시 텍스트를 명시적인 개체-관계 트리플릿으로 변환한다. 이러한 관계 트리플릿은 최종 답변 예측을 위한 간결하고 구조화된 프롬프트로 사용된다. OK-VQA 및 A-OKVQA 벤치마크에서 수행한 광범위한 실험을 통해 SCoRe가 최신 방법들을 일관되게 능가함을 보인다.
In this work, we introduce the Qwen-VL series, a set of large-scale vision-language models (LVLMs) designed to perceive and understand both texts and images. Starting from the Qwen-LM as a foundation, we endow it with visual capacity by the meticulously designed (i) visual receptor, (ii) input-output interface, (iii) 3-stage training pipeline, and (iv) multilingual multimodal cleaned corpus. Beyond the conventional image description and question-answering, we implement the grounding and text-reading ability of Qwen-VLs by aligning image-caption-box tuples. The resulting models, including Qwen-VL and Qwen-VL-Chat, set new records for generalist models under similar model scales on a broad range of visual-centric benchmarks (e.g., image captioning, question answering, visual grounding) and different settings (e.g., zero-shot, few-shot). Moreover, on real-world dialog benchmarks, our instruction-tuned Qwen-VL-Chat also demonstrates superiority compared to existing vision-language chatbots. Code, demo and models are available at https://github.com/QwenLM/Qwen-VL.
Video question answering (VideoQA) is designed to answer a given question based on a relevant video clip. The current available large-scale datasets have made it possible to formulate VideoQA as the joint understanding of visual and language information. However, this training procedure is costly and still less competent with human performance. In this paper, we investigate a transfer learning method by the introduction of domain-agnostic knowledge and domain-specific knowledge. First, we develop a novel transfer learning framework, which finetunes the pre-trained model by applying domain-agnostic knowledge as the medium. Second, we construct a new VideoQA dataset with 21,412 human-generated question-answer samples for comparable transfer of knowledge. Our experiments show that: (i) domain-agnostic knowledge is transferable and (ii) our proposed transfer learning framework can boost VideoQA performance effectively.
Audio question answering (AQA), acting as a widely used proxy task to explore scene understanding, has got more attention. The AQA is challenging for it requires comprehensive temporal reasoning from different scales' events of an audio scene. However, existing methods mostly extend the structures of visual question answering task to audio ones in a simple pattern but may not perform well when perceiving a fine-grained audio scene. To this end, we present a Multi-scale Window Attention Fusion Model (MWAFM) consisting of an asynchronous hybrid attention module and a multi-scale window attention module. The former is designed to aggregate unimodal and cross-modal temporal contexts, while the latter captures sound events of varying lengths and their temporal dependencies for a more comprehensive understanding. Extensive experiments are conducted to demonstrate that the proposed MWAFM can effectively explore temporal information to facilitate AQA in the fine-grained scene.Code: https://github.com/GeWu-Lab/MWAFM
Visual question answering models often produce plausible responses even when the available visual evidence is insufficient to support a valid answer. Although prior work has introduced benchmarks and abstention mechanisms for unanswerable visual questions, it remains unclear which signals derived from a frozen vision-language model reliably indicate answerability on unseen image-question pairs. We systematically evaluate answerability signals on a stratified 10K subset of UNK-VQA using frozen InstructBLIP, covering predictive uncertainty, cross-modal alignment, visual grounding, visual dependence, internal representations, and learned probes. Answerability discrimination remains weak, with test AUROC largely confined to 0.50-0.54. CLIP alignment yields the highest point estimate but does not significantly outperform generation confidence under paired bootstrap analysis. Targeted removal of question-relevant visual evidence causes substantially larger changes in answer confidence than same-size random masking, yet intervention sensitivity provides essentially no answerability discrimination. Increasing probe capacity, combining multiple signals, and adding reason-aware supervision do not provide robust improvements, while post-freeze cross-validation yields similarly weak discrimination. Selective-answering analysis shows limited usable coverage under low-risk operating criteria. Failure-mode analysis shows that the relative utility of individual signals varies across perturbation mechanisms. Taken together, these results show that, on our UNK-VQA subset with frozen InstructBLIP, visual dependence, uncertainty, semantic alignment, and static representations are not robust standalone proxies for answerability. Our findings emphasize the need to distinguish a model's reliance on visual evidence from whether that evidence is sufficient to support an answer.
Building joint representations across images and text is an essential step for tasks such as Visual Question Answering and Video Question Answering. In this work, we find that the representations must not only jointly capture features from both modalities but should also be diverse for better generalization performance. To this end, we propose joint vision-language representation learning by diversifying the tokenization learning process, enabling tokens that are sufficiently disentangled from each other to be learned from both modalities. We observe that our approach outperforms the baseline models in a majority of settings and is competitive with state-of-the-art methods.
There are two main lines of research on visual question answering (VQA): compositional model with explicit multi-hop reasoning, and monolithic network with implicit reasoning in the latent feature space. The former excels in interpretability and compositionality but fails on real-world images, while the latter usually achieves better performance due to model flexibility and parameter efficiency. We aim to combine the two to build an interpretable framework for real-world compositional VQA. In our framework, images and questions are disentangled into scene graphs and programs, and a symbolic program executor runs on them with full transparency to select the attention regions, which are then iteratively passed to a visual-linguistic pre-trained encoder to predict answers. Experiments conducted on the GQA benchmark demonstrate that our framework outperforms the compositional prior arts and achieves competitive accuracy among monolithic ones. With respect to the validity, plausibility and distribution metrics, our framework surpasses others by a considerable margin.
This study focuses on Embodied Complex-Question Answering task, which means the embodied robot need to understand human questions with intricate structures and abstract semantics. The core of this task lies in making appropriate plans based on the perception of the visual environment. Existing methods often generate plans in a once-for-all manner, i.e., one-step planning. Such approach rely on large models, without sufficient understanding of the environment. Considering multi-step planning, the framework for formulating plans in a sequential manner is proposed in this paper. To ensure the ability of our framework to tackle complex questions, we create a structured semantic space, where hierarchical visual perception and chain expression of the question essence can achieve iterative interaction. This space makes sequential task planning possible. Within the framework, we first parse human natural language based on a visual hierarchical scene graph, which can clarify the intention of the question. Then, we incorporate external rules to make a plan for current step, weakening the reliance on large models. Every plan is generated based on feedback from visual perception, with multiple rounds of interaction until an answer is obtained. This approach enables continuous feedback and adjustment, allowing the robot to optimize its action strategy. To test our framework, we contribute a new dataset with more complex questions. Experimental results demonstrate that our approach performs excellently and stably on complex tasks. And also, the feasibility of our approach in real-world scenarios has been established, indicating its practical applicability.
Medical students and junior surgeons often rely on senior surgeons and specialists to answer their questions when learning surgery. However, experts are often busy with clinical and academic work, and have little time to give guidance. Meanwhile, existing deep learning (DL)-based surgical Visual Question Answering (VQA) systems can only provide simple answers without the location of the answers. In addition, vision-language (ViL) embedding is still a less explored research in these kinds of tasks. Therefore, a surgical Visual Question Localized-Answering (VQLA) system would be helpful for medical students and junior surgeons to learn and understand from recorded surgical videos. We propose an end-to-end Transformer with the Co-Attention gaTed Vision-Language (CAT-ViL) embedding for VQLA in surgical scenarios, which does not require feature extraction through detection models. The CAT-ViL embedding module is designed to fuse multimodal features from visual and textual sources. The fused embedding will feed a standard Data-Efficient Image Transformer (DeiT) module, before the parallel classifier and detector for joint prediction. We conduct the experimental validation on public surgical videos from MICCAI EndoVis Challenge 2017 and 2018. The experimental results highlight the superior performance and robustness of our proposed model compared to the state-of-the-art approaches. Ablation studies further prove the outstanding performance of all the proposed components. The proposed method provides a promising solution for surgical scene understanding, and opens up a primary step in the Artificial Intelligence (AI)-based VQLA system for surgical training. Our code is publicly available.
Recent studies have shown that current VQA models are heavily biased on the language priors in the train set to answer the question, irrespective of the image. E.g., overwhelmingly answer "what sport is" as "tennis" or "what color banana" as "yellow." This behavior restricts them from real-world application scenarios. In this work, we propose a novel model-agnostic question encoder, Visually-Grounded Question Encoder (VGQE), for VQA that reduces this effect. VGQE utilizes both visual and language modalities equally while encoding the question. Hence the question representation itself gets sufficient visual-grounding, and thus reduces the dependency of the model on the language priors. We demonstrate the effect of VGQE on three recent VQA models and achieve state-of-the-art results on the bias-sensitive split of the VQAv2 dataset; VQA-CPv2. Further, unlike the existing bias-reduction techniques, on the standard VQAv2 benchmark, our approach does not drop the accuracy; instead, it improves the performance.
Despite the great progress of Visual Question Answering (VQA), current VQA models heavily rely on the superficial correlation between the question type and its corresponding frequent answers (i.e., language priors) to make predictions, without really understanding the input. In this work, we define the training instances with the same question type but different answers as \textit{superficially similar instances}, and attribute the language priors to the confusion of VQA model on such instances. To solve this problem, we propose a novel training framework that explicitly encourages the VQA model to distinguish between the superficially similar instances. Specifically, for each training instance, we first construct a set that contains its superficially similar counterparts. Then we exploit the proposed distinguishing module to increase the distance between the instance and its counterparts in the answer space. In this way, the VQA model is forced to further focus on the other parts of the input beyond the question type, which helps to overcome the language priors. Experimental results show that our method achieves the state-of-the-art performance on VQA-CP v2. Codes are available at \href{https://github.com/wyk-nku/Distinguishing-VQA.git}{Distinguishing-VQA}.
The development of Urdu scene text detection, recognition, and Visual Question Answering (VQA) technologies is crucial for advancing accessibility, information retrieval, and linguistic diversity in digital content, facilitating better understanding and interaction with Urdu-language visual data. This initiative seeks to bridge the gap between textual and visual comprehension. We propose a new multi-task Urdu scene text dataset comprising over 1000 natural scene images, which can be used for text detection, recognition, and VQA tasks. We provide fine-grained annotations for text instances, addressing the limitations of previous datasets for facing arbitrary-shaped texts. By incorporating additional annotation points, this dataset facilitates the development and assessment of methods that can handle diverse text layouts, intricate shapes, and non-standard orientations commonly encountered in real-world scenarios. Besides, the VQA annotations make it the first benchmark for the Urdu Text VQA method, which can prompt the development of Urdu scene text understanding. The proposed dataset is available at: https://github.com/Hiba-MeiRuan/Urdu-VQA-Dataset-/tree/main
Visual attention in Visual Question Answering (VQA) targets at locating the right image regions regarding the answer prediction, offering a powerful technique to promote multi-modal understanding. However, recent studies have pointed out that the highlighted image regions from the visual attention are often irrelevant to the given question and answer, leading to model confusion for correct visual reasoning. To tackle this problem, existing methods mostly resort to aligning the visual attention weights with human attentions. Nevertheless, gathering such human data is laborious and expensive, making it burdensome to adapt well-developed models across datasets. To address this issue, in this paper, we devise a novel visual attention regularization approach, namely AttReg, for better visual grounding in VQA. Specifically, AttReg firstly identifies the image regions which are essential for question answering yet unexpectedly ignored (i.e., assigned with low attention weights) by the backbone model. And then a mask-guided learning scheme is leveraged to regularize the visual attention to focus more on these ignored key regions. The proposed method is very flexible and model-agnostic, which can be integrated into most visual attention-based VQA models and require no human attention supervision. Extensive experiments over three benchmark datasets, i.e., VQA-CP v2, VQA-CP v1, and VQA v2, have been conducted to evaluate the effectiveness of AttReg. As a by-product, when incorporating AttReg into the strong baseline LMH, our approach can achieve a new state-of-the-art accuracy of 60.00% with an absolute performance gain of 7.01% on the VQA-CP v2 benchmark dataset...
RAVEN은 로봇의 시각 임베딩(embedding)을 자세·시간 정보와 함께 벡터 DB에 저장하고 공간 지도에 그라운딩(grounding)해, 캡션 기반 메모리의 정보 손실 없이 장기 질의응답과 목표 지점 탐색을 수행하는 에이전트(agent)형 메모리 시스템이다. 시뮬레이션·실세계 비디오 QA 벤치마크에서 캡션 기반 방식을 일관되게 능가하고, 검색 비용을 10배 낮추면서도 장기 추론 과제에서 최신 VLM 수준에 근접하며, Unitree Go1에서 자연어 목표 장기 내비게이션 배포에도 성공했다.
Electronic Health Records (EHRs), which contain patients' medical histories in various multi-modal formats, often overlook the potential for joint reasoning across imaging and table modalities underexplored in current EHR Question Answering (QA) systems. In this paper, we introduce EHRXQA, a novel multi-modal question answering dataset combining structured EHRs and chest X-ray images. To develop our dataset, we first construct two uni-modal resources: 1) The MIMIC-CXR-VQA dataset, our newly created medical visual question answering (VQA) benchmark, specifically designed to augment the imaging modality in EHR QA, and 2) EHRSQL (MIMIC-IV), a refashioned version of a previously established table-based EHR QA dataset. By integrating these two uni-modal resources, we successfully construct a multi-modal EHR QA dataset that necessitates both uni-modal and cross-modal reasoning. To address the unique challenges of multi-modal questions within EHRs, we propose a NeuralSQL-based strategy equipped with an external VQA API. This pioneering endeavor enhances engagement with multi-modal EHR sources and we believe that our dataset can catalyze advances in real-world medical scenarios such as clinical decision-making and research. EHRXQA is available at https://github.com/baeseongsu/ehrxqa.