NeurIPS 2024 Papers

Large-scale pre-trained models (PTMs) provide remarkable zero-shot classification capability covering a wide variety of object classes. However, practical applications do not always require the classification of all kinds of objects, and leaving the model capable of recognizing unnecessary classes not only degrades overall accuracy but also leads to operational disadvantages. To mitigate this issue, we explore the selective forgetting problem for PTMs, where the task is to make the model unable to recognize only the specified classes, while maintaining accuracy for the rest. All the existing methods assume ''white-box'' settings, where model information such as architectures, parameters, and gradients is available for training. However, PTMs are often ''black-box,'' where information on such models is unavailable for commercial reasons or social responsibilities. In this paper, we address a novel problem of selective forgetting for black-box models, named Black-Box Forgetting, and propose an approach to the problem. Given that information on the model is unavailable, we optimize the input prompt to decrease the accuracy of specified classes through derivative-free optimization. To avoid difficult high-dimensional optimization while ensuring high forgetting performance, we propose Latent Context Sharing, which introduces common low-dimensional latent components among multiple tokens for the prompt. Experiments on four standard benchmark datasets demonstrate the superiority of our method with reasonable baselines. The code is available at https://github.com/yusukekwn/Black-Box-Forgetting.

Understanding documents with rich layouts and multi-modal components is a long-standing and practical task. Recent Large Vision-Language Models (LVLMs) have made remarkable strides in various tasks, particularly in single-page document understanding (DU). However, their abilities on long-context DU remain an open problem. This work presents MMLONGBENCH-DOC, a long-context, multi- modal benchmark comprising 1,082 expert-annotated questions. Distinct from previous datasets, it is constructed upon 135 lengthy PDF-formatted documents with an average of 47.5 pages and 21,214 textual tokens. Towards comprehensive evaluation, answers to these questions rely on pieces of evidence from (1) different sources (text, image, chart, table, and layout structure) and (2) various locations (i.e., page number). Moreover, 33.7\% of the questions are cross-page questions requiring evidence across multiple pages. 20.6\% of the questions are designed to be unanswerable for detecting potential hallucinations. Experiments on 14 LVLMs demonstrate that long-context DU greatly challenges current models. Notably, the best-performing model, GPT-4o, achieves an F1 score of only 44.9\%, while the second-best, GPT-4V, scores 30.5\%. Furthermore, 12 LVLMs (all except GPT-4o and GPT-4V) even present worse performance than their LLM counterparts which are fed with lossy-parsed OCR documents. These results validate the necessity of future research toward more capable long-context LVLMs.

Large vision-language models (LVLMs) are ignorant of the up-to-date knowledge, such as LLaVA series, because they cannot be updated frequently due to the large amount of resources required, and therefore fail in many cases. For example, if a LVLM was released on January 2024, and it wouldn't know the singer of the theme song for the new Detective Conan movie, which wasn't released until April 2024. To solve the problem, a promising solution motivated by retrieval-augmented generation (RAG) is to provide LVLMs with up-to-date knowledge via internet search during inference, i.e., internet-augmented generation (IAG), which is already integrated in some closed-source commercial LVLMs such as GPT-4V. However, the specific mechanics underpinning them remain a mystery. In this paper, we propose a plug-and-play framework, for augmenting existing LVLMs in handling visual question answering (VQA) about up-to-date knowledge, dubbed SearchLVLMs. A hierarchical filtering model is trained to effectively and efficiently find the most helpful content from the websites returned by a search engine to prompt LVLMs with up-to-date knowledge. To train the model and evaluate our framework's performance, we propose a pipeline to automatically generate news-related VQA samples to construct a dataset, dubbed UDK-VQA. A multi-model voting mechanism is introduced to label the usefulness of website/content for VQA samples to construct the training set. Experimental results demonstrate the effectiveness of our framework, outperforming GPT-4o by $\sim$30\% in accuracy.

Speculative decoding has demonstrated its effectiveness in accelerating the inference of large language models (LLMs) while maintaining an identical sampling distribution. However, the conventional approach of training separate draft model to achieve a satisfactory token acceptance rate can be costly and impractical. In this paper, we propose a novel self-speculative decoding framework \emph{Kangaroo} with \emph{double} early exiting strategy, which leverages the shallow sub-network and the \texttt{LM Head} of the well-trained target LLM to construct a self-drafting model. Then, the self-verification stage only requires computing the remaining layers over the \emph{early-exited} hidden states in parallel. To bridge the representation gap between the sub-network and the full model, we train a lightweight and efficient adapter module on top of the sub-network. One significant challenge that comes with the proposed method is that the inference latency of the self-draft model may no longer be negligible compared to the big model. To boost the token acceptance rate while minimizing the latency of the self-drafting model, we introduce an additional \emph{early exiting} mechanism for both single-sequence and the tree decoding scenarios. Specifically, we dynamically halt the small model's subsequent prediction during the drafting phase once the confidence level for the current step falls below a certain threshold. This approach reduces unnecessary computations and improves overall efficiency. Extensive experiments on multiple benchmarks demonstrate our effectiveness, where Kangaroo achieves walltime speedups up to 2.04$\times$, outperforming Medusa-1 with 88.7\% fewer additional parameters. The code for Kangaroo is available at https://github.com/Equationliu/Kangaroo.

Multimodal contrastive learning (MCL) has recently demonstrated significant success across various tasks. However, the existing MCL treats all negative samples equally and ignores the potential semantic association with positive samples, which limits the model's ability to achieve fine-grained alignment. In multi-view scenarios, MCL tends to prioritize shared information while neglecting modality-specific unique information across different views, leading to feature suppression and suboptimal performance in downstream tasks. To address these limitations, we propose a novel contrastive framework name *QUEST: Quadruple Multimodal Contrastive Learning with Constraints and Self-Penalization*. In the QUEST framework, we propose quaternion contrastive objectives and orthogonal constraints to extract sufficient unique information. Meanwhile, a shared information-guided penalization is introduced to ensure that shared information does not excessively influence the optimization of unique information. Our method leverages quaternion vector spaces to simultaneously optimize shared and unique information. Experiments on multiple datasets show that our method achieves superior performance in multimodal contrastive learning benchmarks. On public benchmark, our approach achieves state-of-the-art performance, and on synthetic shortcut datasets, we outperform existing baseline methods by an average of 97.95\% on the CLIP model.

In this paper, we introduce IMPACT (Integrated Multimodal Patent Analysis and Creation Dataset for Design Patents), a large-scale multimodal patent dataset with detailed captions for design patent figures. Our dataset includes half a million design patents comprising 3.61 million figures along with captions from patents granted by the United States Patent and Trademark Office (USPTO) over a 16-year period from 2007 to 2022. We incorporate the metadata of each patent application with elaborate captions that are coherent with multiple viewpoints of designs. Even though patents themselves contain a variety of design figures, titles, and descriptions of viewpoints, we find that they lack detailed descriptions that are necessary to perform multimodal tasks such as classification and retrieval. IMPACT closes this gap thereby providing researchers with necessary ingredients to instantiate a variety of multimodal tasks. Our dataset has a huge potential for novel design inspiration and can be used with advanced computer vision models in tandem. We perform preliminary evaluations on the dataset on the popular patent analysis tasks such as classification and retrieval. Our results indicate that integrating images with generated captions significantly improves the performance of different models on the corresponding tasks. Given that design patents offer various benefits for modeling novel tasks, we propose two standard computer vision tasks that have not been investigated in analyzing patents as future directions using IMPACT as a benchmark viz., 3D Image Construction and Visual Question Answering (VQA). To facilitate research in these directions, we make our IMPACT dataset and the code/models used in this work publicly available at https://github.com/AI4Patents/IMPACT.

We introduce a new benchmark designed to advance the development of general-purpose, large-scale vision-language models for remote sensing images. Although several vision-language datasets in remote sensing have been proposed to pursue this goal, existing datasets are typically tailored to single tasks, lack detailed object information, or suffer from inadequate quality control. Exploring these improvement opportunities, we present a Versatile vision-language Benchmark for Remote Sensing image understanding, termed VRSBench. This benchmark comprises 29,614 images, with 29,614 human-verified detailed captions, 52,472 object references, and 123,221 question-answer pairs. It facilitates the training and evaluation of vision-language models across a broad spectrum of remote sensing image understanding tasks. We further evaluated state-of-the-art models on this benchmark for three vision-language tasks: image captioning, visual grounding, and visual question answering. Our work aims to significantly contribute to the development of advanced vision-language models in the field of remote sensing. The data and code can be accessed at https://vrsbench.github.io.

The swift advancement in Multimodal LLMs (MLLMs) also presents significant challenges for effective knowledge editing. Current methods, including intrinsic knowledge editing and external knowledge resorting, each possess strengths and weaknesses, struggling to balance the desired properties of reliability, generality, and locality when applied to MLLMs. In this paper, we propose \textbf{UniKE}, a novel multimodal editing method that establishes a unified perspective and paradigm for intrinsic knowledge editing and external knowledge resorting. Both types of knowledge are conceptualized as vectorized key-value memories, with the corresponding editing processes resembling the assimilation and accommodation phases of human cognition, conducted at the same semantic levels. Within such a unified framework, we further promote knowledge collaboration by disentangling the knowledge representations into the semantic and truthfulness spaces. Extensive experiments validate the effectiveness of our method, which ensures that the post-edit MLLM simultaneously maintains excellent reliability, generality, and locality. The code for UniKE is available at https://github.com/beepkh/UniKE.

Recent developments of vision large language models (LLMs) have seen remarkable progress, yet still encounter challenges towards multimodal generalists, such as coarse-grained instance-level understanding, lack of unified support for both images and videos, and insufficient coverage across various vision tasks. In this paper we present Vitron, a universal pixel-level vision LLM designed for comprehensive understanding, generating, segmenting, and editing of both static images and dynamic videos. Building on top of an LLM backbone, Vitron incorporates encoders for images, videos, and pixel-level regional visuals within its frontend modules, while employing state-of-the-art visual specialists as its backend, via which Vitron supports a spectrum of vision end tasks, spanning visual comprehension to visual generation, from low level to high level. To ensure an effective and precise message passing from LLM to backend modules for function invocation, we propose a novel hybrid method by simultaneously integrating discrete textual instructions and continuous signal embeddings. Further, we design various pixel-level spatiotemporal vision-language alignment learning for Vitron to reach the best fine-grained visual capability. Finally, a cross-task synergy module is advised to learn to maximize the task-invariant fine-grained visual features, enhancing the synergy between different visual tasks. Demonstrated over 12 visual tasks and evaluated across 22 datasets, Vitron showcases its extensive capabilities in the four main vision task clusters. Overall, this work illuminates the great potential of developing a more unified multimodal generalist.

Visual relationship understanding has been studied separately in human-object interaction(HOI) detection, scene graph generation(SGG), and referring relationships(RR) tasks. Given the complexity and interconnectedness of these tasks, it is crucial to have a flexible framework that can effectively address these tasks in a cohesive manner. In this work, we propose FleVRS, a single model that seamlessly integrates the above three aspects in standard and promptable visual relationship segmentation, and further possesses the capability for open-vocabulary segmentation to adapt to novel scenarios. FleVRS leverages the synergy between text and image modalities, to ground various types of relationships from images and use textual features from vision-language models to visual conceptual understanding. Empirical validation across various datasets demonstrates that our framework outperforms existing models in standard, promptable, and open-vocabulary tasks, e.g., +1.9 $mAP$ on HICO-DET, +11.4 $Acc$ on VRD, +4.7 $mAP$ on unseen HICO-DET. Our FleVRS represents a significant step towards a more intuitive, comprehensive, and scalable understanding of visual relationships.

Image description datasets play a crucial role in the advancement of various applications such as image understanding, text-to-image generation, and text-image retrieval. Currently, image description datasets primarily originate from two sources. One source is the scraping of image-text pairs from the web. Despite their abundance, these descriptions are often of low quality and noisy. Another way is through human labeling. Datasets such as COCO are generally very short and lack details. Although detailed image descriptions can be annotated by humans, the high cost limits their quantity and feasibility. These limitations underscore the need for more efficient and scalable methods to generate accurate and detailed image descriptions. In this paper, we propose an innovative framework termed Image Textualization, which automatically produces high-quality image descriptions by leveraging existing mult-modal large language models (MLLMs) and multiple vision expert models in a collaborative manner. We conduct various experiments to validate the high quality of the descriptions constructed by our framework. Furthermore, we show that MLLMs fine-tuned on our dataset acquire an unprecedented capability of generating richer image descriptions, substantially increasing the length and detail of their output with even less hallucinations.

Powered by remarkable advancements in Large Language Models (LLMs), Multimodal Large Language Models (MLLMs) demonstrate impressive capabilities in manifold tasks. However, the practical application scenarios of MLLMs are intricate, exposing them to potential malicious instructions and thereby posing safety risks. While current benchmarks do incorporate certain safety considerations, they often lack comprehensive coverage and fail to exhibit the necessary rigor and robustness. For instance, the common practice of employing GPT-4V as both the evaluator and a model to be evaluated lacks credibility, as it tends to exhibit a bias toward its own responses. In this paper, we present MLLMGuard, a multi-dimensional safety evaluation suite for MLLMs, including a bilingual image-text evaluation dataset, inference utilities, and a lightweight evaluator. MLLMGuard's assessment comprehensively covers two languages (English and Chinese) and five important safety dimensions (Privacy, Bias, Toxicity, Truthfulness, and Legality), each with corresponding rich subtasks. Focusing on these dimensions, our evaluation dataset is primarily sourced from platforms such as social media, and it integrates text-based and image-based red teaming techniques with meticulous annotation by human experts. This can prevent inaccurate evaluation caused by data leakage when using open-source datasets and ensures the quality and challenging nature of our benchmark. Additionally, a fully automated lightweight evaluator termed GuardRank is developed, which achieves significantly higher evaluation accuracy than GPT-4. Our evaluation results across 13 advanced models indicate that MLLMs still have a substantial journey ahead before they can be considered safe and responsible.

Test-time adaptation, which enables models to generalize to diverse data with unlabeled test samples, holds significant value in real-world scenarios. Recently, researchers have applied this setting to advanced pre-trained vision-language models (VLMs), developing approaches such as test-time prompt tuning to further extend their practical applicability. However, these methods typically focus solely on adapting VLMs from a single modality and fail to accumulate task-specific knowledge as more samples are processed. To address this, we introduce Dual Prototype Evolving (DPE), a novel test-time adaptation approach for VLMs that effectively accumulates task-specific knowledge from multi-modalities. Specifically, we create and evolve two sets of prototypes—textual and visual—to progressively capture more accurate multi-modal representations for target classes during test time. Moreover, to promote consistent multi-modal representations, we introduce and optimize learnable residuals for each test sample to align the prototypes from both modalities. Extensive experimental results on 15 benchmark datasets demonstrate that our proposed DPE consistently outperforms previous state-of-the-art methods while also exhibiting competitive computational efficiency.

As machine learning models in critical fields increasingly grapple with multimodal data, they face the dual challenges of handling a wide array of modalities, often incomplete due to missing elements, and the temporal irregularity and sparsity of collected samples. Successfully leveraging this complex data, while overcoming the scarcity of high-quality training samples, is key to improving these models' predictive performance. We introduce ``FuseMoE'', a mixture-of-experts framework incorporated with an innovative gating function. Designed to integrate a diverse number of modalities, FuseMoE is effective in managing scenarios with missing modalities and irregularly sampled data trajectories. Theoretically, our unique gating function contributes to enhanced convergence rates, leading to better performance in multiple downstream tasks. The practical utility of FuseMoE in the real world is validated by a diverse set of challenging prediction tasks.

We introduce INQUIRE, a text-to-image retrieval benchmark designed to challenge multimodal vision-language models on expert-level queries. INQUIRE includes iNaturalist 2024 (iNat24), a new dataset of five million natural world images, along with 250 expert-level retrieval queries. These queries are paired with all relevant images comprehensively labeled within iNat24, comprising 33,000 total matches. Queries span categories such as species identification, context, behavior, and appearance, emphasizing tasks that require nuanced image understanding and domain expertise. Our benchmark evaluates two core retrieval tasks: (1) INQUIRE-Fullrank, a full dataset ranking task, and (2) INQUIRE-Rerank, a reranking task for refining top-100 retrievals. Detailed evaluation of a range of recent multimodal models demonstrates that INQUIRE poses a significant challenge, with the best models failing to achieve an mAP@50 above 50%. In addition, we show that reranking with more powerful multimodal models can enhance retrieval performance, yet there remains a significant margin for improvement. By focusing on scientifically-motivated ecological challenges, INQUIRE aims to bridge the gap between AI capabilities and the needs of real-world scientific inquiry, encouraging the development of retrieval systems that can assist with accelerating ecological and biodiversity research.

Graphical User Interface (GUI) automation holds significant promise for enhancing human productivity by assisting with computer tasks. Existing task formulations primarily focus on simple tasks that can be specified by a single, language-only instruction, such as “Insert a new slide.” In this work, we introduce VideoGUI, a novel multi-modal benchmark designed to evaluate GUI assistants on visual-centric GUI tasks. Sourced from high-quality web instructional videos, our benchmark focuses on tasks involving professional and novel software (e.g., Adobe Pho- toshop or Stable Diffusion WebUI) and complex activities (e.g., video editing). VideoGUI evaluates GUI assistants through a hierarchical process, allowing for identification of the specific levels at which they may fail: (i) high-level planning: reconstruct procedural subtasks from visual conditions without language descrip- tions; (ii) middle-level planning: generate sequences of precise action narrations based on visual state (i.e., screenshot) and goals; (iii) atomic action execution: perform specific actions such as accurately clicking designated elements. For each level, we design evaluation metrics across individual dimensions to provide clear signals, such as individual performance in clicking, dragging, typing, and scrolling for atomic action execution. Our evaluation on VideoGUI reveals that even the SoTA large multimodal model GPT4o performs poorly on visual-centric GUI tasks, especially for high-level planning. The data and code are available at https://github.com/showlab/videogui.

Large Language Models (LLMs) have demonstrated impressive capabilities across a range of scientific tasks including mathematics, physics, and chemistry. Despite their successes, the effectiveness of LLMs in handling complex statistical tasks remains systematically under-explored. To bridge this gap, we introduce StatQA, a new benchmark designed for statistical analysis tasks. StatQA comprises 11,623 examples tailored to evaluate LLMs' proficiency in specialized statistical tasks and their applicability assessment capabilities, particularly for hypothesis testing methods. We systematically experiment with representative LLMs using various prompting strategies and show that even state-of-the-art models such as GPT-4o achieve a best performance of only 64.83%, indicating significant room for improvement. Notably, while open-source LLMs (e.g. LLaMA-3) show limited capability, those fine-tuned ones exhibit marked improvements, outperforming all in-context learning-based methods (e.g. GPT-4o). Moreover, our comparative human experiments highlight a striking contrast in error types between LLMs and humans: LLMs primarily make applicability errors, whereas humans mostly make statistical task confusion errors. This divergence highlights distinct areas of proficiency and deficiency, suggesting that combining LLM and human expertise could lead to complementary strengths, inviting further investigation into their collaborative potential. Our source code and data are available at https://statqa.github.io/.

Large Language Models (LLMs) are being deployed across various domains today. However, their capacity to solve Capture the Flag (CTF) challenges in cybersecurity has not been thoroughly evaluated. To address this, we develop a novel method to assess LLMs in solving CTF challenges by creating a scalable, open-source benchmark database specifically designed for these applications. This database includes metadata for LLM testing and adaptive learning, compiling a diverse range of CTF challenges from popular competitions. Utilizing the advanced function calling capabilities of LLMs, we build a fully automated system with an enhanced workflow and support for external tool calls. Our benchmark dataset and automated framework allow us to evaluate the performance of five LLMs, encompassing both black-box and open-source models. This work lays the foundation for future research into improving the efficiency of LLMs in interactive cybersecurity tasks and automated task planning. By providing a specialized benchmark, our project offers an ideal platform for developing, testing, and refining LLM-based approaches to vulnerability detection and resolution. Evaluating LLMs on these challenges and comparing with human performance yields insights into their potential for AI-driven cybersecurity solutions to perform real-world threat management. We make our benchmark dataset open source to public https://github.com/NYU-LLM-CTF/NYU_CTF_Bench along with our playground automated framework https://github.com/NYU-LLM-CTF/llm_ctf_automation.

Recent advances in multi-modal algorithms have driven and been driven by the increasing availability of large image-text datasets, leading to significant strides in various fields, including computational pathology. However, in most existing medical image-text datasets, the text typically provides high-level summaries that may not sufficiently describe sub-tile regions within a large pathology image. For example, an image might cover an extensive tissue area containing cancerous and healthy regions, but the accompanying text might only specify that this image is a cancer slide, lacking the nuanced details needed for in-depth analysis. In this study, we introduce STimage-1K4M, a novel dataset designed to bridge this gap by providing genomic features for sub-tile images. STimage-1K4M contains 1,149 images derived from spatial transcriptomics data, which captures gene expression information at the level of individual spatial spots within a pathology image. Specifically, each image in the dataset is broken down into smaller sub-image tiles, with each tile paired with $15,000-30,000$ dimensional gene expressions. With $4,293,195$ pairs of sub-tile images and gene expressions, STimage-1K4M offers unprecedented granularity, paving the way for a wide range of advanced research in multi-modal data analysis an innovative applications in computational pathology, and beyond.