PyTorchKR Papers

최신 논문

총 1,987편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2610.15728v2cs.CV

When Watermarks Turn Adversarial: Exposing Vulnerabilities in Document Understanding

Visible watermarks are common in digital documents and persist during vision-language model (VLM) processing. We show that they can also serve as adversarial surfaces for document understanding. We propose Watermark-based Adversarial Surface Perturbation (WASP), a white-box attack that optimizes pixels inside fixed watermark strokes without requiring downstream questions, answers, or task-specific losses. We further introduce Balanced Aggregation Loss (BAL), which emphasizes visual token groups that remain highly similar to clean representations. Adversarial watermarks degrade performance substantially more than ordinary watermarks. On DocVQA, WASP achieves attack success rates of 96.81% and 71.93% against Qwen2.5-VL and InternVL2.5; BAL improves over a standard token hinge objective by 9.21 and 2.24 percentage points. On 1,651 OmniDocBench pages, WASP increases MinerU2.5's text edit distance by 0.7748 relative to ordinary watermarks. WASP also transfers within model families and across readers sharing a vision-tower architecture. Attack effectiveness depends on watermark geometry, model architecture, and preprocessing; in structured parsing, disrupted layout and routing can propagate errors downstream. These results expose an underexplored security risk in document understanding systems.

arXiv:2510.04532v3cs.AI

보이는 것 그 이상인가? 비전-언어 주행 모델 학습에서 추론과 계획 간의 단절 규명

More Than Meets the Eye? Uncovering the Reasoning-Planning Disconnect in Training Vision-Language Driving Models

비전-언어 모델(VLM) 기반 주행 에이전트는 자연어 추론을 먼저 생성한 뒤 궤적 계획을 예측함으로써 설명 가능한 엔드투엔드 자율주행을 실현할 것으로 기대받고 있습니다. 그러나 계획이 이러한 추론에 의해 인과적으로 구동되는지는 매우 중요하지만 검증되지 않은 가정으로 남아 있습니다. 이를 조사하기 위해, 본 연구에서는 nuPlan으로부터 자동 생성된 계획 정렬 사고의 연쇄(CoT)를 포함하는 대규모 주행 시각적 질의응답(VQA) 코퍼스인 DriveMind를 구축합니다. 본 데이터 생성 과정은 센서와 어노테이션을 정형화된 입력으로 변환하며, 특히 사전 정보(prior)와 추론 대상 신호를 분리하여 명확한 정보 절제(ablation)를 가능하게 합니다. DriveMind를 사용하여 지도 미세조정(SFT) 및 그룹 상대 정책 최적화(GRPO)로 대표적인 VLM 에이전트를 학습시키고 nuPlan 메트릭으로 평가합니다. 안타깝게도 실험 결과는 추론-계획 간의 일관된 인과적 단절을 나타냅니다. 자차(ego)/내비게이션 사전 정보를 제거하면 계획 점수가 크게 하락하는 반면, CoT를 제거해도 점수 변화는 미미합니다. 어텐션 분석 또한 계획 단계가 CoT보다는 주로 사전 정보에 집중함을 보여줍니다. 이러한 증거를 바탕으로, 본 연구는 학습을 통해 산출된 추론이 인과적 매개체가 아니라 부수적인 부산물에 불과하다고 상정하는 '추론-계획 분리 가설(Reasoning-Planning Decoupling Hypothesis)'을 제안합니다. 효율적인 진단을 위해, 사소한 입력 섭동에 대한 계획 강건성을 평가하여 에이전트의 사전 정보 의존도를 측정하는 새로운 학습 무관(training-free) 프로브를 도입합니다. 요약하자면, 본 연구는 향후 모델의 인과적 충실성을 평가할 수 있는 새로운 데이터셋과 진단 도구를 커뮤니티에 제공합니다.

arXiv:2507.13348cs.CV

VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning

VisionThink은 시각-언어 모델(VLM)의 시각 토큰 수를 동적으로 조절하여 효율성을 높이는 새로운 방법론입니다. 초기에는 저해상도 이미지를 사용하고, 문제 해결에 충분하지 않을 경우에만 고해상도 이미지를 요청하는 방식으로, 고정된 토큰 압축 비율 대신 샘플별로 최적의 압축을 수행합니다. 강화학습과 LLM-as-Judge 전략을 도입하여 일반적인 시각질문응답(VQA) 과제에 적용하였으며, 보상 함수와 페널티 메커니즘을 통해 안정적인 이미지 크기 조절을 달성하였습니다. 실험 결과, OCR 관련 과제에서는 세밀한 시각 이해 능력을 유지하면서도 단순 과제에서는 시각 토큰 수를 크게 절감하는 우수한 성능과 효율성을 입증하였습니다.

arXiv:2507.12441공식 코드 · Papers with Code

Describe Anything Model for Visual Question Answering on Text-rich Images

Recent progress has been made in region-aware vision-language modeling, particularly with the emergence of the Describe Anything Model (DAM). DAM is capable of generating detailed descriptions of any specific image areas or objects without the need for additional localized image-text alignment supervision. We hypothesize that such region-level descriptive capability is beneficial for the task of Visual Question Answering (VQA), especially in challenging scenarios involving images with dense text. In such settings, the fine-grained extraction of textual information is crucial to producing correct answers. Motivated by this, we introduce DAM-QA, a framework with a tailored evaluation protocol, developed to investigate and harness the region-aware capabilities from DAM for the text-rich VQA problem that requires reasoning over text-based information within images. DAM-QA incorporates a mechanism that aggregates answers from multiple regional views of image content, enabling more effective identification of evidence that may be tied to text-related elements. Experiments on six VQA benchmarks show that our approach consistently outperforms the baseline DAM, with a notable 7+ point gain on DocVQA. DAM-QA also achieves the best overall performance among region-aware models with fewer parameters, significantly narrowing the gap with strong generalist VLMs. These results highlight the potential of DAM-like models for text-rich and broader VQA tasks when paired with efficient usage and integration strategies. Our code is publicly available at https://github.com/Linvyl/DAM-QA.git.

arXiv:2507.12232

MGFFD-VLM: Multi-Granularity Prompt Learning for Face Forgery Detection with VLM

Recent studies have utilized visual large language models (VLMs) to answer not only "Is this face a forgery?" but also "Why is the face a forgery?" These studies introduced forgery-related attributes, such as forgery location and type, to construct deepfake VQA datasets and train VLMs, achieving high accuracy while providing human-understandable explanatory text descriptions. However, these methods still have limitations. For example, they do not fully leverage face quality-related attributes, which are often abnormal in forged faces, and they lack effective training strategies for forgery-aware VLMs. In this paper, we extend the VQA dataset to create DD-VQA+, which features a richer set of attributes and a more diverse range of samples. Furthermore, we introduce a novel forgery detection framework, MGFFD-VLM, which integrates an Attribute-Driven Hybrid LoRA Strategy to enhance the capabilities of Visual Large Language Models (VLMs). Additionally, our framework incorporates Multi-Granularity Prompt Learning and a Forgery-Aware Training Strategy. By transforming classification and forgery segmentation results into prompts, our method not only improves forgery classification but also enhances interpretability. To further boost detection performance, we design multiple forgery-related auxiliary losses. Experimental results demonstrate that our approach surpasses existing methods in both text-based forgery judgment and analysis, achieving superior accuracy.

arXiv:2506.17939

GEMeX-ThinkVG: Towards Thinking with Visual Grounding in Medical VQA via Reinforcement Learning

Medical visual question answering aims to support clinical decision-making by enabling models to answer natural language questions based on medical images. While recent advances in multi-modal learning have significantly improved performance, current methods still suffer from limited answer reliability and poor interpretability, impairing the ability of clinicians and patients to understand and trust model-generated answers. To address this, this work first proposes a Thinking with Visual Grounding (ThinkVG) dataset wherein the answer generation is decomposed into intermediate reasoning steps that explicitly ground relevant visual regions of the medical image, thereby providing fine-grained explainability. Furthermore, we introduce a novel verifiable reward mechanism for reinforcement learning to guide post-training, improving the alignment between the model's reasoning process and its final answer. Remarkably, our method achieves comparable performance using only one-eighth of the training data, demonstrating the efficiency and effectiveness of the proposal. The dataset is available at https://huggingface.co/datasets/BoKelvin/GEMeX-ThinkVG.

arXiv:2507.07274

LinguaMark: Do Multimodal Models Speak Fairly? A Benchmark-Based Evaluation

Large Multimodal Models (LMMs) are typically trained on vast corpora of image-text data but are often limited in linguistic coverage, leading to biased and unfair outputs across languages. While prior work has explored multimodal evaluation, less emphasis has been placed on assessing multilingual capabilities. In this work, we introduce LinguaMark, a benchmark designed to evaluate state-of-the-art LMMs on a multilingual Visual Question Answering (VQA) task. Our dataset comprises 6,875 image-text pairs spanning 11 languages and five social attributes. We evaluate models using three key metrics: Bias, Answer Relevancy, and Faithfulness. Our findings reveal that closed-source models generally achieve the highest overall performance. Both closed-source (GPT-4o and Gemini2.5) and open-source models (Gemma3, Qwen2.5) perform competitively across social attributes, and Qwen2.5 demonstrates strong generalization across multiple languages. We release our benchmark and evaluation code to encourage reproducibility and further research.

arXiv:2507.07079

Evaluating Attribute Confusion in Fashion Text-to-Image Generation

Despite the rapid advances in Text-to-Image (T2I) generation models, their evaluation remains challenging in domains like fashion, involving complex compositional generation. Recent automated T2I evaluation methods leverage pre-trained vision-language models to measure cross-modal alignment. However, our preliminary study reveals that they are still limited in assessing rich entity-attribute semantics, facing challenges in attribute confusion, i.e., when attributes are correctly depicted but associated to the wrong entities. To address this, we build on a Visual Question Answering (VQA) localization strategy targeting one single entity at a time across both visual and textual modalities. We propose a localized human evaluation protocol and introduce a novel automatic metric, Localized VQAScore (L-VQAScore), that combines visual localization with VQA probing both correct (reflection) and miss-localized (leakage) attribute generation. On a newly curated dataset featuring challenging compositional alignment scenarios, L-VQAScore outperforms state-of-the-art T2I evaluation methods in terms of correlation with human judgments, demonstrating its strength in capturing fine-grained entity-attribute associations. We believe L-VQAScore can be a reliable and scalable alternative to subjective evaluations.

arXiv:2506.21710

FOCUS: Internal MLLM Representations for Efficient Fine-Grained Visual Question Answering

While Multimodal Large Language Models (MLLMs) offer strong perception and reasoning capabilities for image-text input, Visual Question Answering (VQA) focusing on small image details still remains a challenge. Although visual cropping techniques seem promising, recent approaches have several limitations: the need for task-specific fine-tuning, low efficiency due to uninformed exhaustive search, or incompatibility with efficient attention implementations. We address these shortcomings by proposing a training-free visual cropping method, dubbed FOCUS, that leverages MLLM-internal representations to guide the search for the most relevant image region. This is accomplished in four steps: first, we identify the target object(s) in the VQA prompt; second, we compute an object relevance map using the key-value (KV) cache; third, we propose and rank relevant image regions based on the map; and finally, we perform the fine-grained VQA task using the topranked region. As a result of this informed search strategy, FOCUS achieves strong performance across four fine-grained VQA datasets and two types of MLLMs. It outperforms three popular visual cropping methods in both accuracy and efficiency, and matches the best-performing baseline, ZoomEye, while requiring 3 – 6.5× less compute.

arXiv:2506.22880공식 코드 · Papers with Code

Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and Grounder

Existing video segmenter and grounder approaches, exemplified by Sa2VA, directly fuse features within segmentation models. This often results in an undesirable entanglement of dynamic visual information and static semantics, thereby degrading segmentation accuracy. To systematically mitigate this issue, we propose DeSa2VA, a decoupling-enhanced prompting scheme integrating text pre-training and a linear decoupling module to address the information processing limitations inherent in SAM-2. Specifically, first, we devise a pre-training paradigm that converts textual ground-truth labels into point-level prompts while generating corresponding text masks. These masks are refined through a hybrid loss function to strengthen the model's semantic grounding capabilities. Next, we employ linear projection to disentangle hidden states that generated by a large language model into distinct textual and visual feature subspaces. Finally, a dynamic mask fusion strategy synergistically combines these decoupled features through triple supervision from predicted text/visual masks and ground-truth annotations. Extensive experiments demonstrate state-of-the-art performance across diverse tasks, including image segmentation, image question answering, video segmentation, and video question answering. Our codes are available at https://github.com/longmalongma/DeSa2VA.

arXiv:2502.00843

VLM-Assisted Continual learning for Visual Question Answering in Self-Driving

In this paper, we propose a novel approach for solving the Visual Question Answering (VQA) task in autonomous driving by integrating Vision-Language Models (VLMs) with continual learning. In autonomous driving, VQA plays a vital role in enabling the system to understand and reason about its surroundings. However, traditional models often struggle with catastrophic forgetting when sequentially exposed to new driving tasks, such as perception, prediction, and planning, each requiring different forms of knowledge. To address this challenge, we present a novel continual learning framework that combines VLMs with selective memory replay and knowledge distillation, reinforced by task-specific projection layer regularization. The knowledge distillation allows a previously trained model to act as a "teacher" to guide the model through subsequent tasks, minimizing forgetting. Meanwhile, task-specific projection layers calculate the loss based on the divergence of feature representations, ensuring continuity in learning and reducing the shift between tasks. Evaluated on the DriveLM dataset, our framework shows substantial performance improvements, with gains ranging from 21.40% to 32.28% across various metrics. These results highlight the effectiveness of combining continual learning with VLMs in enhancing the resilience and reliability of VQA systems in autonomous driving. We will release our source code.

arXiv:2412.00151공식 코드 · Papers with Code

DLaVA: Document Language and Vision Assistant for Answer Localization with Enhanced Interpretability and Trustworthiness

Document Visual Question Answering (VQA) requires models to interpret textual information within complex visual layouts and comprehend spatial relationships to answer questions based on document images. Existing approaches often lack interpretability and fail to precisely localize answers within the document, hindering users' ability to verify responses and understand the reasoning process. Moreover, standard metrics like Average Normalized Levenshtein Similarity (ANLS) focus on text accuracy but overlook spatial correctness. We introduce DLaVA, a novel method that enhances Multimodal Large Language Models (MLLMs) with answer localization capabilities for Document VQA. Our approach integrates image annotation directly into the MLLM pipeline, improving interpretability by enabling users to trace the model's reasoning. We present both OCR-dependent and OCR-free architectures, with the OCR-free approach eliminating the need for separate text recognition components, thus reducing complexity. To the best of our knowledge, DLaVA is the first approach to introduce answer localization within multimodal QA, marking a significant step forward in enhancing user trust and reducing the risk of AI hallucinations. Our contributions include enhancing interpretability and reliability by grounding responses in spatially annotated visual content, introducing answer localization in MLLMs, proposing a streamlined pipeline that combines an MLLM with a text detection module, and conducting comprehensive evaluations using both textual and spatial accuracy metrics, including Intersection over Union (IoU). Experimental results on standard datasets demonstrate that DLaVA achieves SOTA performance, significantly enhancing model transparency and reliability. Our approach sets a new benchmark for Document VQA, highlighting the critical importance of precise answer localization and model interpretability.

arXiv:2506.16450

How Far Can Off-the-Shelf Multimodal Large Language Models Go in Online Episodic Memory Question Answering?

We investigate whether off-the-shelf Multimodal Large Language Models (MLLMs) can tackle Online Episodic-Memory Video Question Answering (OEM-VQA) without additional training. Our pipeline converts a streaming egocentric video into a lightweight textual memory, only a few kilobytes per minute, via an MLLM descriptor module, and answers multiple-choice questions by querying this memory with an LLM reasoner module. On the QAEgo4D-Closed benchmark, our best configuration attains 56.0% accuracy with 3.6 kB per minute storage, matching the performance of dedicated state-of-the-art systems while being 10**4/10**5 times more memory-efficient. Extensive ablations provides insights into the role of each component and design choice, and highlight directions of improvement for future research.

arXiv:2506.21355

SMMILE: An Expert-Driven Benchmark for Multimodal Medical In-Context Learning

Multimodal in-context learning (ICL) remains underexplored despite significant potential for domains such as medicine. Clinicians routinely encounter diverse, specialized tasks requiring adaptation from limited examples, such as drawing insights from a few relevant prior cases or considering a constrained set of differential diagnoses. While multimodal large language models (MLLMs) have shown advances in medical visual question answering (VQA), their ability to learn multimodal tasks from context is largely unknown. We introduce SMMILE, the first expert-driven multimodal ICL benchmark for medical tasks. Eleven medical experts curated problems, each including a multimodal query and multimodal in-context examples as task demonstrations. SMMILE encompasses 111 problems (517 question-image-answer triplets) covering 6 medical specialties and 13 imaging modalities. We further introduce SMMILE++, an augmented variant with 1038 permuted problems. A comprehensive evaluation of 15 MLLMs demonstrates that most models exhibit moderate to poor multimodal ICL ability in medical tasks. In open-ended evaluations, ICL contributes only 8% average improvement over zero-shot on SMMILE and 9.4% on SMMILE++. We observe a susceptibility for irrelevant in-context examples: even a single noisy or irrelevant example can degrade performance by up to 9.5%. Moreover, example ordering exhibits a recency bias, i.e., placing the most relevant example last can lead to substantial performance improvements by up to 71%. Our findings highlight critical limitations and biases in current MLLMs when learning multimodal medical tasks from context.

arXiv:2506.21316공식 코드 · Papers with Code

DrishtiKon: Multi-Granular Visual Grounding for Text-Rich Document Images

Visual grounding in text-rich document images is a critical yet underexplored challenge for document intelligence and visual question answering (VQA) systems. We present \drishtikon, a multi-granular visual grounding framework designed to enhance interpretability and trust in VQA for complex, multilingual documents. Our approach integrates robust multi-lingual OCR, large language models, and a novel region matching algorithm to accurately localize answer spans at block, line, word, and point levels. We curate a new benchmark from the CircularsVQA test set, providing fine-grained, human-verified annotations across multiple granularities. Extensive experiments demonstrate that our method achieves state-of-the-art grounding accuracy, with line-level granularity offering the best trade-off between precision and recall. Ablation studies further highlight the benefits of multi-block and multi-line reasoning. Comparative evaluations with leading vision-language models reveal the limitations of current VLMs in precise localization, underscoring the effectiveness of our structured, alignment-based approach. Our findings pave the way for more robust and interpretable document understanding systems in real-world, text-centric scenarios. Code and dataset has been made available at https://github.com/kasuba-badri-vishal/DhrishtiKon.

arXiv:2506.21011

Bridging Video Quality Scoring and Justification via Large Multimodal Models

Classical video quality assessment (VQA) methods generate a numerical score to judge a video's perceived visual fidelity and clarity. Yet, a score fails to describe the video's complex quality dimensions, restricting its applicability. Benefiting from the linguistic output, adapting video large multimodal models (LMMs) to VQA via instruction tuning has the potential to address this issue. The core of the approach lies in the video quality-centric instruction data. Previous explorations mainly focus on the image domain, and their data generation processes heavily rely on human quality annotations and proprietary systems, limiting data scalability and effectiveness. To address these challenges, we propose the Score-based Instruction Generation (SIG) pipeline. Specifically, SIG first scores multiple quality dimensions of an unlabeled video and maps scores to text-defined levels. It then explicitly incorporates a hierarchical Chain-of-Thought (CoT) to model the correlation between specific dimensions and overall quality, mimicking the human visual system's reasoning process. The automated pipeline eliminates the reliance on expert-written quality descriptions and proprietary systems, ensuring data scalability and generation efficiency. To this end, the resulting Score2Instruct (S2I) dataset contains over 320K diverse instruction-response pairs, laying the basis for instruction tuning. Moreover, to advance video LMMs' quality scoring and justification abilities simultaneously, we devise a progressive tuning strategy to fully unleash the power of S2I. Built upon SIG, we further curate a benchmark termed S2I-Bench with 400 open-ended questions to better evaluate the quality justification capacity of video LMMs. Experimental results on the S2I-Bench and existing benchmarks indicate that our method consistently improves quality scoring and justification capabilities across multiple video LMMs.

arXiv:2506.11549

EyeSim-VQA: A Free-Energy-Guided Eye Simulation Framework for Video Quality Assessment

Free-energy-guided self-repair mechanisms have shown promising results in image quality assessment (IQA), but remain under-explored in video quality assessment (VQA), where temporal dynamics and model constraints pose unique challenges. Unlike static images, video content exhibits richer spatiotemporal complexity, making perceptual restoration more difficult. Moreover, VQA systems often rely on pre-trained backbones, which limits the direct integration of enhancement modules without affecting model stability. To address these issues, we propose EyeSimVQA, a novel VQA framework that incorporates free-energy-based self-repair. It adopts a dual-branch architecture, with an aesthetic branch for global perceptual evaluation and a technical branch for fine-grained structural and semantic analysis. Each branch integrates specialized enhancement modules tailored to distinct visual inputs-resized full-frame images and patch-based fragments-to simulate adaptive repair behaviors. We also explore a principled strategy for incorporating high-level visual features without disrupting the original backbone. In addition, we design a biologically inspired prediction head that models sweeping gaze dynamics to better fuse global and local representations for quality prediction. Experiments on five public VQA benchmarks demonstrate that EyeSimVQA achieves competitive or superior performance compared to state-of-the-art methods, while offering improved interpretability through its biologically grounded design.

arXiv:2506.15298

MEGC2025: Micro-Expression Grand Challenge on Spot Then Recognize and Visual Question Answering

Facial micro-expressions (MEs) are involuntary movements of the face that occur spontaneously when a person experiences an emotion but attempts to suppress or repress the facial expression, typically found in a high-stakes environment. In recent years, substantial advancements have been made in the areas of ME recognition, spotting, and generation. However, conventional approaches that treat spotting and recognition as separate tasks are suboptimal, particularly for analyzing long-duration videos in realistic settings. Concurrently, the emergence of multimodal large language models (MLLMs) and large vision-language models (LVLMs) offers promising new avenues for enhancing ME analysis through their powerful multimodal reasoning capabilities. The ME grand challenge (MEGC) 2025 introduces two tasks that reflect these evolving research directions: (1) ME spot-then-recognize (ME-STR), which integrates ME spotting and subsequent recognition in a unified sequential pipeline; and (2) ME visual question answering (ME-VQA), which explores ME understanding through visual question answering, leveraging MLLMs or LVLMs to address diverse question types related to MEs. All participating algorithms are required to run on this test set and submit their results on a leaderboard. More details are available at https://megc2025.github.io.

arXiv:2506.12849

CAPO: Reinforcing Consistent Reasoning in Medical Decision-Making

In medical visual question answering (Med-VQA), achieving accurate responses relies on three critical steps: precise perception of medical imaging data, logical reasoning grounded in visual input and textual questions, and coherent answer derivation from the reasoning process. Recent advances in general vision-language models (VLMs) show that large-scale reinforcement learning (RL) could significantly enhance both reasoning capabilities and overall model performance. However, their application in medical domains is hindered by two fundamental challenges: 1) misalignment between perceptual understanding and reasoning stages, and 2) inconsistency between reasoning pathways and answer generation, both compounded by the scarcity of high-quality medical datasets for effective large-scale RL. In this paper, we first introduce Med-Zero-17K, a curated dataset for pure RL-based training, encompassing over 30 medical image modalities and 24 clinical tasks. Moreover, we propose a novel large-scale RL framework for Med-VLMs, Consistency-Aware Preference Optimization (CAPO), which integrates rewards to ensure fidelity between perception and reasoning, consistency in reasoning-to-answer derivation, and rule-based accuracy for final responses. Extensive experiments on both in-domain and out-of-domain scenarios demonstrate the superiority of our method over strong VLM baselines, showcasing strong generalization capability to 3D Med-VQA benchmarks and R1-like training paradigms.

arXiv:2505.21755공식 코드 · Papers with Code

FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering

Visual question answering (VQA) systems face significant challenges when adapting to real-world data shifts, especially in multi-modal contexts. While robust fine-tuning strategies are essential for maintaining performance across in-distribution (ID) and out-of-distribution (OOD) scenarios, current evaluation settings are primarily unimodal or particular to some types of OOD, offering limited insight into the complexities of multi-modal contexts. In this work, we propose a new benchmark FRAMES-VQA (Fine-Tuning Robustness across Multi-Modal Shifts in VQA) for evaluating robust fine-tuning for VQA tasks. We utilize ten existing VQA benchmarks, including VQAv2, IV-VQA, VQA-CP, OK-VQA and others, and categorize them into ID, near and far OOD datasets covering uni-modal, multi-modal and adversarial distribution shifts. We first conduct a comprehensive comparison of existing robust fine-tuning methods. We then quantify the distribution shifts by calculating the Mahalanobis distance using uni-modal and multi-modal embeddings extracted from various models. Further, we perform an extensive analysis to explore the interactions between uni- and multi-modal shifts as well as modality importance for ID and OOD samples. These analyses offer valuable guidance on developing more robust fine-tuning methods to handle multi-modal distribution shifts. The code is available at https://github.com/chengyuehuang511/FRAMES-VQA .