PyTorchKR Papers

“Visual Question Answering”

총 2,804편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2204.14198

Flamingo: a Visual Language Model for Few-Shot Learning

Building models that can be rapidly adapted to novel tasks using only a handful of annotated examples is an open challenge for multimodal machine learning research. We introduce Flamingo, a family of Visual Language Models (VLM) with this ability. We propose key architectural innovations to: (i) bridge powerful pretrained vision-only and language-only models, (ii) handle sequences of arbitrarily interleaved visual and textual data, and (iii) seamlessly ingest images or videos as inputs. Thanks to their flexibility, Flamingo models can be trained on large-scale multimodal web corpora containing arbitrarily interleaved text and images, which is key to endow them with in-context few-shot learning capabilities. We perform a thorough evaluation of our models, exploring and measuring their ability to rapidly adapt to a variety of image and video tasks. These include open-ended tasks such as visual question-answering, where the model is prompted with a question which it has to answer; captioning tasks, which evaluate the ability to describe a scene or an event; and close-ended tasks such as multiple-choice visual question-answering. For tasks lying anywhere on this spectrum, a single Flamingo model can achieve a new state of the art with few-shot learning, simply by prompting the model with task-specific examples. On numerous benchmarks, Flamingo outperforms models fine-tuned on thousands of times more task-specific data.

arXiv:2401.15842

LCV2: An Efficient Pretraining-Free Framework for Grounded Visual Question Answering

In this paper, the LCV2 modular method is proposed for the Grounded Visual Question Answering task in the vision-language multimodal domain. This approach relies on a frozen large language model (LLM) as intermediate mediator between the off-the-shelf VQA model and the off-the-shelf visual grounding (VG) model, where the LLM transforms and conveys textual information between the two modules based on a designed prompt. LCV2 establish an integrated plug-and-play framework without the need for any pre-training process. This framework can be deployed for VQA Grounding tasks under low computational resources. The modularized model within the framework allows application with various state-of-the-art pre-trained models, exhibiting significant potential to be advance with the times. Experimental implementations were conducted under constrained computational and memory resources, evaluating the proposed method's performance on benchmark datasets including GQA, CLEVR, and VizWiz-VQA-Grounding. Comparative analyses with baseline methods demonstrate the robust competitiveness of LCV2.

arXiv:2305.14882

Dynamic Clue Bottlenecks: Towards Interpretable-by-Design Visual Question Answering

Recent advances in multimodal large language models (LLMs) have shown extreme effectiveness in visual question answering (VQA). However, the design nature of these end-to-end models prevents them from being interpretable to humans, undermining trust and applicability in critical domains. While post-hoc rationales offer certain insight into understanding model behavior, these explanations are not guaranteed to be faithful to the model. In this paper, we address these shortcomings by introducing an interpretable by design model that factors model decisions into intermediate human-legible explanations, and allows people to easily understand why a model fails or succeeds. We propose the Dynamic Clue Bottleneck Model ( (DCLUB), a method that is designed towards an inherently interpretable VQA system. DCLUB provides an explainable intermediate space before the VQA decision and is faithful from the beginning, while maintaining comparable performance to black-box systems. Given a question, DCLUB first returns a set of visual clues: natural language statements of visually salient evidence from the image, and then generates the output based solely on the visual clues. To supervise and evaluate the generation of VQA explanations within DCLUB, we collect a dataset of 1.7k reasoning-focused questions with visual clues. Evaluations show that our inherently interpretable system can improve 4.64% over a comparable black-box system in reasoning-focused questions while preserving 99.43% of performance on VQA-v2.

arXiv:2611.12938v1cs.AI

지식 그래프 질의응답을 위한 프롬프트 내 그래프 시각화

Graph Visualizations in Prompts for Knowledge Graph Question Answering

본 연구는 대규모 시각-언어 모델에서 멀티모달 입력이 지식 그래프 질의응답(KGQA)을 언제, 어떻게 향상시키는지 연구한다. 고정된 KGQA 파이프라인 내에서 WebQSP와 CWQ의 1,500개 질문과 여섯 개의 VLM을 대상으로 통제된 평가를 수행하여, 질문 단위로 텍스트 전용 프롬프팅과 텍스트에 그래프 시각화를 더한 프롬프팅을 비교하고, 추론 복잡도, 모델 크기, 이미지 레이아웃에 따른 성능 변화를 분석한다. 그래프 이미지를 추가하면 평균 1.85 F1 포인트의 향상이 나타나며, 이는 낮은 복잡도의 질문에서 나타난 큰 정밀도 향상(+4.65 F1)에 기인한다. 혼합효과 분석과 회귀 분석 결과, 이러한 멀티모달 이득은 모델 규모에 크게 의존하지 않으나 그래프 레이아웃에는 매우 민감하다. 이미지만 사용하는 설정은 성능을 크게 저하시키고, 계층적 Graphviz 레이아웃은 덜 정돈된 렌더링보다 우수하며, 큰 부분 그래프를 분리된 경로로 분해하면 가장 큰 향상을 얻는다. 종합하면, 시각적 프롬프팅은 시각 설계가 인지적·지각적 제약을 존중하는 경우 낮은 복잡도의 상황에서 예측 정밀도를 높임으로써 주로 KGQA를 돕는다는 점을 결과가 시사한다. 정확도와 추론 효율을 모두 최적화하기 위해, 잘 구조화된 레이아웃을 가진 작고 시각적으로 성긴 부분 그래프에만 선택적으로 도식을 덧붙이고 복잡한 질의에서는 불필요한 시각적 혼잡과 토큰 오버헤드를 피하는 적응형 시각 프롬프팅 정책을 제안한다. 마지막으로 이러한 발견을 시각적 표기법 설계 이론과 연관 지어, 원리에 입각한 도식 인코딩이 향후 멀티모달 VLM 프롬프팅을 어떻게 이끌 수 있는지 서술한다.

arXiv:2404.12372

MedThink: Explaining Medical Visual Question Answering via Multimodal Decision-Making Rationale

Medical Visual Question Answering (MedVQA), which offers language responses to image-based medical inquiries, represents a challenging task and significant advancement in healthcare. It assists medical experts to swiftly interpret medical images, thereby enabling faster and more accurate diagnoses. However, the model interpretability and transparency of existing MedVQA solutions are often limited, posing challenges in understanding their decision-making processes. To address this issue, we devise a semi-automated annotation process to streamline data preparation and build new benchmark MedVQA datasets R-RAD, R-SLAKE and R-Path. These datasets provide intermediate medical decision-making rationales generated by multimodal large language models and human annotations for question-answering pairs in existing MedVQA datasets, i.e., VQA-RAD, SLAKE and PathVQA. Moreover, we design a novel framework, MedThink, which finetunes lightweight pretrained generative models by incorporating medical decision-making rationales. MedThink includes three distinct strategies to generate decision outcomes and corresponding rationales, thereby clearly showcasing the medical decision-making process during reasoning. Our comprehensive experiments show that our method achieves an accuracy of 83.5% on R-RAD, 86.3% on R-SLAKE and 87.2% on R-Path. These results significantly exceed those of existing state-of-the-art models with comparable parameters. Datasets and code will be released.

arXiv:2405.10948

Surgical-LVLM: Learning to Adapt Large Vision-Language Model for Grounded Visual Question Answering in Robotic Surgery

Recent advancements in Surgical Visual Question Answering (Surgical-VQA) and related region grounding have shown great promise for robotic and medical applications, addressing the critical need for automated methods in personalized surgical mentorship. However, existing models primarily provide simple structured answers and struggle with complex scenarios due to their limited capability in recognizing long-range dependencies and aligning multimodal information. In this paper, we introduce Surgical-LVLM, a novel personalized large vision-language model tailored for complex surgical scenarios. Leveraging the pre-trained large vision-language model and specialized Visual Perception LoRA (VP-LoRA) blocks, our model excels in understanding complex visual-language tasks within surgical contexts. In addressing the visual grounding task, we propose the Token-Interaction (TIT) module, which strengthens the interaction between the grounding module and the language responses of the Large Visual Language Model (LVLM) after projecting them into the latent space. We demonstrate the effectiveness of Surgical-LVLM on several benchmarks, including EndoVis-17-VQLA, EndoVis-18-VQLA, and a newly introduced EndoVis Conversations dataset, which sets new performance standards. Our work contributes to advancing the field of automated surgical mentorship by providing a context-aware solution.

arXiv:2611.10642v1cs.CV

VisProve: 과학 문서 질의응답을 위한 유형화된 시각적 출처의 학습

VisProve: Learning Typed Visual Provenance for Scientific Document Question Answering

시각적 인용은 답이 어디에서 뒷받침되는지를 보여주지만, 인용된 내용이 답에 어떻게 기여하는지는 보여주지 않는다. 과학 문서 질의응답에서 모델은 값을 직접 읽을 수도, 여러 관측을 압축할 수도, 증거를 결합하여 결론을 추론할 수도 있다. 우리는 시스템이 답과 함께 근거 영역 및 명시적인 지지 관계를 생성하는 유형화된 시각적 출처(typed visual provenance)를 연구한다. 우리는 완전한 과학 문서에 대한 3,941개의 질문으로 구성되고, 페이지, 섹션, 영역 주석이 인용(Quotation), 압축(Compression), 추론(Inference)으로 레이블링된 데이터셋 ReViE를 소개한다. 나아가 미분 가능한 공간 증거 읽기(Spatial Evidence Reading)와 증거-답 관계 피드백(Evidence-Answer Relation Feedback)으로 답과 출처의 공동 생성을 강화하는 VisProve를 제안한다. VisProve는 게이트가 적용된 노름 제한 잔차를 통해 다음 토큰 예측을 국소화된 시각적 증거와 그 추정된 지지 관계에 조건화한다. 문서가 서로 겹치지 않는 ReViE 테스트 세트에서 VisProve는 동일한 백본의 공동 지도 미세조정 대비 Typed-F1을 3.52에서 5.48로, Section-F1을 45.99에서 53.24로 향상시키는 반면 Page-F1은 근소하게만 변한다. 이러한 결과는 시각적 근거화와 지지 관계 모델링을 결합하면 주로 세밀한 출처가 개선됨을 시사하며, 낮은 절대 Typed-F1은 답 생성과 정밀한 시각적 증거 귀속 사이에 상당한 격차가 있음을 보여준다.

arXiv:2501.02201

Accounting for Focus Ambiguity in Visual Questions

No existing work on visual question answering explicitly accounts for ambiguity regarding where the content described in the question is located in the image. To fill this gap, we introduce VQ-FocusAmbiguity, the first VQA dataset that visually grounds each region described in the question that is necessary to arrive at the answer. We then provide an analysis showing how our dataset for visually grounding `questions' is distinct from visually grounding `answers', and characterize the properties of the questions and segmentations provided in our dataset. Finally, we benchmark modern models for two novel tasks: recognizing whether a visual question has focus ambiguity and localizing all plausible focus regions within the image. Results show that the dataset is challenging for modern models. To facilitate future progress on these tasks, we publicly share the dataset with an evaluation server at https://focusambiguity.github.io/.

arXiv:2407.02352

Pelican: Correcting Hallucination in Vision-LLMs via Claim Decomposition and Program of Thought Verification

Large Visual Language Models (LVLMs) struggle with hallucinations in visual instruction following task(s), limiting their trustworthiness and real-world applicability. We propose Pelican -- a novel framework designed to detect and mitigate hallucinations through claim verification. Pelican first decomposes the visual claim into a chain of sub-claims based on first-order predicates. These sub-claims consist of (predicate, question) pairs and can be conceptualized as nodes of a computational graph. We then use Program-of-Thought prompting to generate Python code for answering these questions through flexible composition of external tools. Pelican improves over prior work by introducing (1) intermediate variables for precise grounding of object instances, and (2) shared computation for answering the sub-question to enable adaptive corrections and inconsistency identification. We finally use reasoning abilities of LLMs to verify the correctness of the claim by considering the consistency and confidence of the (question, answer) pairs from each sub-claim. Our experiments reveal a drop in hallucination rate by ~ 8% - 32% across various baseline LVLMs and a 27% drop compared to approaches proposed for hallucination mitigation on MMHal-Bench. Results on two other benchmarks further corroborate our results.

arXiv:2304.06061공식 코드 · Papers with Code

CLIP-Guided Vision-Language Pre-training for Question Answering in 3D Scenes

Training models to apply linguistic knowledge and visual concepts from 2D images to 3D world understanding is a promising direction that researchers have only recently started to explore. In this work, we design a novel 3D pre-training Vision-Language method that helps a model learn semantically meaningful and transferable 3D scene point cloud representations. We inject the representational power of the popular CLIP model into our 3D encoder by aligning the encoded 3D scene features with the corresponding 2D image and text embeddings produced by CLIP. To assess our model's 3D world reasoning capability, we evaluate it on the downstream task of 3D Visual Question Answering. Experimental quantitative and qualitative results show that our pre-training method outperforms state-of-the-art works in this task and leads to an interpretable representation of 3D scene features.

arXiv:2310.17914공식 코드 · Papers with Code

3D-Aware Visual Question Answering about Parts, Poses and Occlusions

Despite rapid progress in Visual question answering (VQA), existing datasets and models mainly focus on testing reasoning in 2D. However, it is important that VQA models also understand the 3D structure of visual scenes, for example to support tasks like navigation or manipulation. This includes an understanding of the 3D object pose, their parts and occlusions. In this work, we introduce the task of 3D-aware VQA, which focuses on challenging questions that require a compositional reasoning over the 3D structure of visual scenes. We address 3D-aware VQA from both the dataset and the model perspective. First, we introduce Super-CLEVR-3D, a compositional reasoning dataset that contains questions about object parts, their 3D poses, and occlusions. Second, we propose PO3D-VQA, a 3D-aware VQA model that marries two powerful ideas: probabilistic neural symbolic program execution for reasoning and deep neural networks with 3D generative representations of objects for robust visual recognition. Our experimental results show our model PO3D-VQA outperforms existing methods significantly, but we still observe a significant performance gap compared to 2D VQA benchmarks, indicating that 3D-aware VQA remains an important open research area.

arXiv:1809.01943공식 코드 · Papers with Code

Cascaded Mutual Modulation for Visual Reasoning

Visual reasoning is a special visual question answering problem that is multi-step and compositional by nature, and also requires intensive text-vision interactions. We propose CMM: Cascaded Mutual Modulation as a novel end-to-end visual reasoning model. CMM includes a multi-step comprehension process for both question and image. In each step, we use a Feature-wise Linear Modulation (FiLM) technique to enable textual/visual pipeline to mutually control each other. Experiments show that CMM significantly outperforms most related models, and reach state-of-the-arts on two visual reasoning benchmarks: CLEVR and NLVR, collected from both synthetic and natural languages. Ablation studies confirm that both our multistep framework and our visual-guided language modulation are critical to the task. Our code is available at https://github.com/FlamingHorizon/CMM-VR.

arXiv:2404.19205공식 코드 · Papers with Code

TableVQA-Bench: A Visual Question Answering Benchmark on Multiple Table Domains

In this paper, we establish a benchmark for table visual question answering, referred to as the TableVQA-Bench, derived from pre-existing table question-answering (QA) and table structure recognition datasets. It is important to note that existing datasets have not incorporated images or QA pairs, which are two crucial components of TableVQA. As such, the primary objective of this paper is to obtain these necessary components. Specifically, images are sourced either through the application of a \textit{stylesheet} or by employing the proposed table rendering system. QA pairs are generated by exploiting the large language model (LLM) where the input is a text-formatted table. Ultimately, the completed TableVQA-Bench comprises 1,500 QA pairs. We comprehensively compare the performance of various multi-modal large language models (MLLMs) on TableVQA-Bench. GPT-4V achieves the highest accuracy among commercial and open-sourced MLLMs from our experiments. Moreover, we discover that the number of vision queries plays a significant role in TableVQA performance. To further analyze the capabilities of MLLMs in comparison to their LLM backbones, we investigate by presenting image-formatted tables to MLLMs and text-formatted tables to LLMs, respectively. Our findings suggest that processing visual inputs is more challenging than text inputs, as evidenced by the lower performance of MLLMs, despite generally requiring higher computational costs than LLMs. The proposed TableVQA-Bench and evaluation codes are available at \href{https://github.com/naver-ai/tablevqabench}{https://github.com/naver-ai/tablevqabench}.

arXiv:2002.05104

Component Analysis for Visual Question Answering Architectures

Recent research advances in Computer Vision and Natural Language Processing have introduced novel tasks that are paving the way for solving AI-complete problems. One of those tasks is called Visual Question Answering (VQA). A VQA system must take an image and a free-form, open-ended natural language question about the image, and produce a natural language answer as the output. Such a task has drawn great attention from the scientific community, which generated a plethora of approaches that aim to improve the VQA predictive accuracy. Most of them comprise three major components: (i) independent representation learning of images and questions; (ii) feature fusion so the model can use information from both sources to answer visual questions; and (iii) the generation of the correct answer in natural language. With so many approaches being recently introduced, it became unclear the real contribution of each component for the ultimate performance of the model. The main goal of this paper is to provide a comprehensive analysis regarding the impact of each component in VQA models. Our extensive set of experiments cover both visual and textual elements, as well as the combination of these representations in form of fusion and attention mechanisms. Our major contribution is to identify core components for training VQA models so as to maximize their predictive performance.

arXiv:2005.00330

Visuo-Linguistic Question Answering (VLQA) Challenge

Understanding images and text together is an important aspect of cognition and building advanced Artificial Intelligence (AI) systems. As a community, we have achieved good benchmarks over language and vision domains separately, however joint reasoning is still a challenge for state-of-the-art computer vision and natural language processing (NLP) systems. We propose a novel task to derive joint inference about a given image-text modality and compile the Visuo-Linguistic Question Answering (VLQA) challenge corpus in a question answering setting. Each dataset item consists of an image and a reading passage, where questions are designed to combine both visual and textual information i.e., ignoring either modality would make the question unanswerable. We first explore the best existing vision-language architectures to solve VLQA subsets and show that they are unable to reason well. We then develop a modular method with slightly better baseline performance, but it is still far behind human performance. We believe that VLQA will be a good benchmark for reasoning over a visuo-linguistic context. The dataset, code and leaderboard is available at https://shailaja183.github.io/vlqa/.

arXiv:2610.00757v1cs.CV

비디오 증거 인덱싱(Video Evidence Indexing): 토큰 예산 제약 장시간 비디오 질의응답을 위해 비디오 프리뷰로부터 어디를 볼지 학습하기

Video Evidence Indexing: Learning Where to Look from Video Previews for Token-Budgeted Long-Video Question Answering

장시간 비디오 질의응답(long-video question answering)은 비주얼 토큰의 높은 비용과 현재 VLM(비전-언어 모델)의 고정된 컨텍스트 폭(context width)으로 인해 제약을 받는다. 장시간 비디오에 대한 질문은 넓은 시간적 범위(temporal coverage)를 요구할 수 있지만, 정답은 흔히 소수의 압축된 순간들(moments)에 의해서만 뒷받침된다. 이러한 순간들을 효율적으로 찾아내기 위해, 우리는 토큰 예산이 제한된 비디오 증거 인덱싱(token-budgeted Video Evidence Indexing, VEI)을 제안한다: 조밀한 저해상도 비디오 프리뷰(Video Preview)가 주어지면, 모델은 최종 추론을 위한 압축된 고해상도 증거 집합(Evidence Set)을 구성한다. 우리는 VEI를 질문과 관련된 순간을 찾는 증거 위치 추정(evidence localization)과, 제한된 고해상도 프레임 예산을 어디에 사용할지 결정하는 예산 계획(budget planning)을 함께 해결해야 하는 정책(policy)으로 다룬다. 우리는 이 아이디어를 하나의 추론 파이프라인으로 구현한다: 비디오 프리뷰는 저비용의 전역적 커버리지를 제공하고, 비디오 증거 인덱싱은 증거 집합을 구성하며, 답변 생성(Answer Generation)은 최종 VQA를 위해 두 입력을 결합한다. 프레임 수준의 감독(supervision)이 부재한 문제를 해결하기 위해, 우리는 특권적 자기 증류(privileged self-distillation)를 채택하는데, 여기서는 정답을 아는 교사(answer-aware teacher)가 학생이 생성한 인덱싱 흔적(indexing trace)에 대해 일반적인 테스트 시점(test-time) 정책을 지도한다. 우리는 프레임당 1, 6, 12, 24개의 비주얼 토큰을 가진 프리뷰를 탐구하며, 이 네 가지 해상도를 모두 지원하는 단일 정책을 학습시킨다. 실험 결과, 비디오 증거 인덱싱은 제한된 비주얼 예산 하에서 정확도를 향상시키며, 자기 증류는 QA 정확도와 시간적 증거 위치 추정(temporal evidence localization) 모두를 추가로 향상시킨다.

arXiv:2301.10799

Towards a Unified Model for Generating Answers and Explanations in Visual Question Answering

The field of visual question answering (VQA) has recently seen a surge in research focused on providing explanations for predicted answers. However, current systems mostly rely on separate models to predict answers and generate explanations, leading to less grounded and frequently inconsistent results. To address this, we propose a multitask learning approach towards a Unified Model for Answer and Explanation generation (UMAE). Our approach involves the addition of artificial prompt tokens to training data and fine-tuning a multimodal encoder-decoder model on a variety of VQA-related tasks. In our experiments, UMAE models surpass the prior state-of-the-art answer accuracy on A-OKVQA by 10~15%, show competitive results on OK-VQA, achieve new state-of-the-art explanation scores on A-OKVQA and VCR, and demonstrate promising out-of-domain performance on VQA-X.

arXiv:2306.00758

LiT-4-RSVQA: Lightweight Transformer-based Visual Question Answering in Remote Sensing

Visual question answering (VQA) methods in remote sensing (RS) aim to answer natural language questions with respect to an RS image. Most of the existing methods require a large amount of computational resources, which limits their application in operational scenarios in RS. To address this issue, in this paper we present an effective lightweight transformer-based VQA in RS (LiT-4-RSVQA) architecture for efficient and accurate VQA in RS. Our architecture consists of: i) a lightweight text encoder module; ii) a lightweight image encoder module; iii) a fusion module; and iv) a classification module. The experimental results obtained on a VQA benchmark dataset demonstrate that our proposed LiT-4-RSVQA architecture provides accurate VQA results while significantly reducing the computational requirements on the executing hardware. Our code is publicly available at https://git.tu-berlin.de/rsim/lit4rsvqa.

arXiv:2006.05726

Estimating semantic structure for the VQA answer space

Since its appearance, Visual Question Answering (VQA, i.e. answering a question posed over an image), has always been treated as a classification problem over a set of predefined answers. Despite its convenience, this classification approach poorly reflects the semantics of the problem limiting the answering to a choice between independent proposals, without taking into account the similarity between them (e.g. equally penalizing for answering cat or German shepherd instead of dog). We address this issue by proposing (1) two measures of proximity between VQA classes, and (2) a corresponding loss which takes into account the estimated proximity. This significantly improves the generalization of VQA models by reducing their language bias. In particular, we show that our approach is completely model-agnostic since it allows consistent improvements with three different VQA models. Finally, by combining our method with a language bias reduction approach, we report SOTA-level performance on the challenging VQAv2-CP dataset.

arXiv:1707.07998공식 코드 · Papers with Code

Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

Top-down visual attention mechanisms have been used extensively in image captioning and visual question answering (VQA) to enable deeper image understanding through fine-grained analysis and even multiple steps of reasoning. In this work, we propose a combined bottom-up and top-down attention mechanism that enables attention to be calculated at the level of objects and other salient image regions. This is the natural basis for attention to be considered. Within our approach, the bottom-up mechanism (based on Faster R-CNN) proposes image regions, each with an associated feature vector, while the top-down mechanism determines feature weightings. Applying this approach to image captioning, our results on the MSCOCO test server establish a new state-of-the-art for the task, achieving CIDEr / SPICE / BLEU-4 scores of 117.9, 21.5 and 36.9, respectively. Demonstrating the broad applicability of the method, applying the same approach to VQA we obtain first place in the 2017 VQA Challenge.