NeurIPS 2024 Papers

Open-vocabulary 3D object detection (OV-3Det) aims to generalize beyond the limited number of base categories labeled during the training phase. The biggest bottleneck is the scarcity of annotated 3D data, whereas 2D image datasets are abundant and richly annotated. Consequently, it is intuitive to leverage the wealth of annotations in 2D images to alleviate the inherent data scarcity in OV-3Det. In this paper, we push the task setup to its limits by exploring the potential of using solely 2D images to learn OV-3Det. The major challenges for this setup is the modality gap between training images and testing point clouds, which prevents effective integration of 2D knowledge into OV-3Det. To address this challenge, we propose a novel framework ImOV3D to leverage pseudo multimodal representation containing both images and point clouds (PC) to close the modality gap. The key of ImOV3D lies in flexible modality conversion where 2D images can be lifted into 3D using monocular depth estimation and can also be derived from 3D scenes through rendering. This allows unifying both training images and testing point clouds into a common image-PC representation, encompassing a wealth of 2D semantic information and also incorporating the depth and structural characteristics of 3D spatial data. We carefully conduct such conversion to minimize the domain gap between training and test cases. Extensive experiments on two benchmark datasets, SUNRGBD and ScanNet, show that ImOV3D significantly outperforms existing methods, even in the absence of ground truth 3D training data. With the inclusion of a minimal amount of real 3D data for fine-tuning, the performance also significantly surpasses previous state-of-the-art. Codes and pre-trained models are released on the https://github.com/yangtiming/ImOV3D.

Detecting Human-Object Interactions (HOI) in zero-shot settings, where models must handle unseen classes, poses significant challenges. Existing methods that rely on aligning visual encoders with large Vision-Language Models (VLMs) to tap into the extensive knowledge of VLMs, require large, computationally expensive models and encounter training difficulties. Adapting VLMs with prompt learning offers an alternative to direct alignment. However, fine-tuning on task-specific datasets often leads to overfitting to seen classes and suboptimal performance on unseen classes, due to the absence of unseen class labels. To address these challenges, we introduce a novel prompt learning-based framework for Efficient Zero-Shot HOI detection (EZ-HOI). First, we introduce Large Language Model (LLM) and VLM guidance for learnable prompts, integrating detailed HOI descriptions and visual semantics to adapt VLMs to HOI tasks. However, because training datasets contain seen-class labels alone, fine-tuning VLMs on such datasets tends to optimize learnable prompts for seen classes instead of unseen ones. Therefore, we design prompt learning for unseen classes using information from related seen classes, with LLMs utilized to highlight the differences between unseen and related seen classes. Quantitative evaluations on benchmark datasets demonstrate that our EZ-HOI achieves state-of-the-art performance across various zero-shot settings with only 10.35\% to 33.95\% of the trainable parameters compared to existing methods. Code is available at https://github.com/ChelsieLei/EZ-HOI.

Data science and engineering workflows often span multiple stages, from warehousing to orchestration, using tools like BigQuery, dbt, and Airbyte. As vision language models (VLMs) advance in multimodal understanding and code generation, VLM-based agents could potentially automate these workflows by generating SQL queries, Python code, and GUI operations. This automation can improve the productivity of experts while democratizing access to large-scale data analysis. In this paper, we introduce Spider2-V, the first multimodal agent benchmark focusing on professional data science and engineering workflows, featuring 494 real-world tasks in authentic computer environments and incorporating 20 enterprise-level professional applications. These tasks, derived from real-world use cases, evaluate the ability of a multimodal agent to perform data-related tasks by writing code and managing the GUI in enterprise data software systems. To balance realistic simulation with evaluation simplicity, we devote significant effort to developing automatic configurations for task setup and carefully crafting evaluation metrics for each task. Furthermore, we supplement multimodal agents with comprehensive documents of these enterprise data software systems. Our empirical evaluation reveals that existing state-of-the-art LLM/VLM-based agents do not reliably automate full data workflows (14.0% success). Even with step-by-step guidance, these agents still underperform in tasks that require fine-grained, knowledge-intensive GUI actions (16.2%) and involve remote cloud-hosted workspaces (10.6%). We hope that Spider2-V paves the way for autonomous multimodal agents to transform the automation of data science and engineering workflow. Our code and data are available at https://spider2-v.github.io.

As language models continue to scale, Large Language Models (LLMs) have exhibited emerging capabilities in In-Context Learning (ICL), enabling them to solve language tasks by prefixing a few in-context demonstrations (ICDs) as context. Inspired by these advancements, researchers have extended these techniques to develop Large Multimodal Models (LMMs) with ICL capabilities. However, applying ICL usually faces two major challenges: 1) using more ICDs will largely increase the inference time and 2) the performance is sensitive to the selection of ICDs. These challenges are further exacerbated in LMMs due to the integration of multiple data types and the combinational complexity of multimodal ICDs. Recently, to address these challenges, some NLP studies introduce non-learnable In-Context Vectors (ICVs) which extract useful task information from ICDs into a single vector and then insert it into the LLM to help solve the corresponding task. However, although useful in simple NLP tasks, these non-learnable methods fail to handle complex multimodal tasks like Visual Question Answering (VQA). In this study, we propose \underline{\textbf{L}}earnable \underline{\textbf{I}}n-Context \underline{\textbf{Ve}}ctor (LIVE) to distill essential task information from demonstrations, improving ICL performance in LMMs. Experiments show that LIVE can significantly reduce computational costs while enhancing accuracy in VQA tasks compared to traditional ICL and other non-learnable ICV methods.

The advent of large vision-language models (LVLMs) has spurred research into their applications in multi-modal contexts, particularly in video understanding. Traditional VideoQA benchmarks, despite providing quantitative metrics, often fail to encompass the full spectrum of video content and inadequately assess models' temporal comprehension. To address these limitations, we introduce MMBench-Video, a quantitative benchmark designed to rigorously evaluate LVLMs' proficiency in video understanding. MMBench-Video incorporates lengthy videos from YouTube and employs free-form questions, mirroring practical use cases. The benchmark is meticulously crafted to probe the models' temporal reasoning skills, with all questions human-annotated according to a carefully constructed ability taxonomy. We employ GPT-4 for automated assessment, demonstrating superior accuracy and robustness over earlier LLM-based evaluations. Utilizing MMBench-Video, we have conducted comprehensive evaluations that include both proprietary and open-source LVLMs for images and videos. MMBench-Video stands as a valuable resource for the research community, facilitating improved evaluation of LVLMs and catalyzing progress in the field of video understanding.

Multimodal learning falls into the trap of the optimization dilemma due to the modality imbalance phenomenon, leading to unsatisfactory performance in real applications. A core reason for modality imbalance is that the models of each modality converge at different rates. Many attempts naturally focus on adjusting learning procedures adaptively. Essentially, the reason why models converge at different rates is because the difficulty of fitting category labels is inconsistent for each modality during learning. From the perspective of fitting labels, we find that appropriate positive intervention label fitting can correct this difference in learning ability. By exploiting the ability of contrastive learning to intervene in the learning of category label fitting, we propose a novel multimodal learning approach that dynamically integrates unsupervised contrastive learning and supervised multimodal learning to address the modality imbalance problem. We find that a simple yet heuristic integration strategy can significantly alleviate the modality imbalance phenomenon. Moreover, we design a learning-based integration strategy to integrate two losses dynamically, further improving the performance. Experiments on widely used datasets demonstrate the superiority of our method compared with state-of-the-art (SOTA) multimodal learning approaches. The code is available at https://github.com/njustkmg/NeurIPS24-LFM.

Recent advancements in Large Multimodal Models (LMMs) have shown promising results in mathematical reasoning within visual contexts, with models exceeding human-level performance on existing benchmarks such as MathVista. However, we observe significant limitations in the diversity of questions and breadth of subjects covered by these benchmarks. To address this issue, we present the MATH-Vision (MATH-V) dataset, a meticulously curated collection of 3,040 high-quality mathematical problems with visual contexts sourced from real math competitions. Spanning 16 distinct mathematical disciplines and graded across 5 levels of difficulty, our dataset provides a comprehensive and diverse set of challenges for evaluating the mathematical reasoning abilities of LMMs. Through extensive experimentation, we unveil a notable performance gap between current LMMs and human performance on \datasetname, underscoring the imperative for further advancements in LMMs. Moreover, our detailed categorization allows for a thorough error analysis of LMMs, offering valuable insights to guide future research and development. The dataset is released at [MathLLMs/MathVision](https://huggingface.co/datasets/MathLLMs/MathVision)

Instruction tuning (IT) is crucial to tailoring large language models (LLMs) towards human-centric interactions. Recent advancements have shown that the careful selection of a small, high-quality subset of IT data can significantly enhance the performance of LLMs. Despite this, common approaches often rely on additional models or data, which increases costs and limits widespread adoption. In this work, we propose a novel approach, termed $\textit{SelectIT}$, that capitalizes on the foundational capabilities of the LLM itself. Specifically, we exploit the intrinsic uncertainty present in LLMs to more effectively select high-quality IT data, without the need for extra resources. Furthermore, we introduce a curated IT dataset, the $\textit{Selective Alpaca}$, created by applying SelectIT to the Alpaca-GPT4 dataset. Empirical results demonstrate that IT using Selective Alpaca leads to substantial model ability enhancement. The robustness of SelectIT has also been corroborated in various foundation models and domain-specific tasks. Our findings suggest that longer and more computationally intensive IT data may serve as superior sources of IT, offering valuable insights for future research in this area. Data, code, and scripts are freely available at https://github.com/Blue-Raincoat/SelectIT.

Visual content and accompanied audio signals naturally formulate a joint representation to improve audio-visual (AV) related applications. While studies develop various AV representation learning frameworks, the importance of AV data alignment is usually undermined for achieving high-quality representation. We observe that an audio signal may contain background noise interference. Also, non-synchronization may appear between audio and video streams. These non-strict data alignment limits representation quality and downgrade application performance. In this paper, we propose to improve AV joint representations from a data-centric perspective by aligning audio signals to visual data. Our alignment is conducted in an agentic workflow controlled by an LLM-based assistant named AVAgent. For each input AV data pair, our AVAgent uses a multi-modal LLM to convert audio and visual data into language descriptions separately (i.e., tool use). Then, AVAgent reasons whether this paired data is aligned well and plans to edit the audio signal if needed (i.e., planning). The audio editing is executed by predefined actions that filter noise or augment data. Moreover, we use a VLM to evaluate how modified audio signals match the visual content and provide feedback to AVAgent (i.e., reflection). The tool use, planning, and reflection steps operate cyclically to become an agentic workflow where audio signals are gradually aligned to visual content. To this end, existing methods can directly leverage the aligned AV data via our agentic workflow to improve AV joint representations. The experimental results comprehensively demonstrate the state-of-the-art performance of the proposed approach against previous baselines in diverse downstream tasks.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable comprehension and reasoning capabilities with complex language and visual data. These advances have spurred the vision of establishing a generalist robotic MLLM proficient in understanding complex human instructions and accomplishing various embodied tasks, whose feasibility has been recently verified~\cite{rt-2,rt-x}. However, developing MLLMs for real-world robots is challenging due to the typically limited computation and memory capacities available on robotic platforms. In contrast, the inference of MLLMs usually incorporates storing billions of parameters and performing tremendous computation, imposing significant hardware demands. In our paper, we seek to address this challenge by leveraging an intriguing observation: relatively easier situations make up the bulk of the procedure of controlling robots to fulfill diverse tasks, and they generally require far smaller models to obtain the correct robotic actions. Motivated by this observation, we propose a \emph{Dynamic Early-Exit for Robotic MLLM} (DeeR) framework that automatically adjusts the size of the activated MLLM based on each situation at hand. The approach leverages a multi-exit architecture in MLLMs, which allows the model to cease processing once a proper size of the model has been activated for a specific situation, thus avoiding further redundant computation. Additionally, we develop novel algorithms that establish early-termination criteria for DeeR, conditioned on predefined demands such as average computational cost (\emph{i.e.}, power consumption), as well as peak computational consumption (\emph{i.e.}, latency) and GPU memory usage. These enhancements ensure that DeeR operates efficiently under varying resource constraints while maintaining competitive performance. Moreover, we design a tailored training method for integrating temporal information on top of such multi-exit architectures to predict actions reasonably. On the CALVIN robot manipulation benchmark, DeeR demonstrates significant reductions in computational costs by 5.2-6.5x and GPU memory by 2x without compromising performance. Code and checkpoints are available at https://github.com/yueyang130/DeeR-VLA.

Multi-turn visual conversation is an important ability of real-world AI assistants. However, the related evaluation benchmark is missed. This paper presents ConvBench, a multi-turn conversation benchmark with hierarchical capabilities ablation evaluation for Large Vision-Language Models (LVLMs). ConvBench comprises 577 curated multi-turn conversations, encompassing 215 tasks. These tasks are broad and open-ended, which resemble real-world user behaviors. ConvBench progressively examines the LVLMs' perception, reasoning, and creativity capabilities in each conversation and can decouple these capabilities in evaluations and thus perform reliable error attribution. Besides, considering the diversity of open-ended questions, we introduce an efficient and reliable automatic evaluation framework. Experimental results reveal that ConvBench is a significant challenge for current LVLMs, even for GPT4V, which achieves only a 39.51% score. Besides, we have some insightful findings, such as the weak perception of LVLMs inhibits authentic strengths in reasoning and creation. We believe our design of hierarchical capabilities, decoupling capabilities evaluation, and multi-turn conversation can blaze a new trail in LVLMs evaluation. Code and benchmark are released at https://github.com/shirlyliu64/ConvBench.

Severe data imbalance naturally exists among web-scale vision-language datasets. Despite this, we find CLIP pre-trained thereupon exhibits notable robustness to the data imbalance compared to supervised learning, and demonstrates significant effectiveness in learning generalizable representations. With an aim to investigate the reasons behind this finding, we conduct controlled experiments to study various underlying factors, and reveal that CLIP's pretext task forms a dynamic classification problem wherein only a subset of classes is present in training. This isolates the bias from dominant classes and implicitly balances the learning signal. Furthermore, the robustness and discriminability of CLIP improve with more descriptive language supervision, larger data scale, and broader open-world concepts, which are inaccessible to supervised learning. Our study not only uncovers the mechanisms behind CLIP's generalizability beyond data imbalance but also provides transferable insights for the research community. The findings are validated in both supervised and self-supervised learning, enabling models trained on imbalanced data to achieve CLIP-level performance on diverse recognition tasks. Code and data are available at: https://github.com/CVMI-Lab/clip-beyond-tail.

Video understanding relies on accurate action detection for temporal analysis. However, existing mainstream methods have limitations in real-world applications due to their offline and closed-set evaluation approaches, as well as their dependence on manual annotations. To address these challenges and enable real-time action understanding in open-world scenarios, we propose OV-OAD, a zero-shot online action detector that leverages vision-language models and learns solely from text supervision. By introducing an object-centered decoder unit into a Transformer-based model, we aggregate frames with similar semantics using video-text correspondence. Extensive experiments on four action detection benchmarks demonstrate that OV-OAD outperforms other advanced zero-shot methods. Specifically, it achieves 37.5\% mean average precision on THUMOS’14 and 73.8\% calibrated average precision on TVSeries. This research establishes a robust baseline for zero-shot transfer in online action detection, enabling scalable solutions for open-world temporal understanding. The code will be available for download at \url{https://github.com/OpenGVLab/OV-OAD}.

We present a novel OCR-free document understanding framework based on pretrained Multimodal Large Language Models (MLLMs). Our approach employs multi-scale visual features to effectively handle various font sizes within document images. To address the increasing costs of considering the multi-scale visual inputs for MLLMs, we propose the Hierarchical Visual Feature Aggregation (HVFA) module, designed to reduce the number of input tokens to LLMs. Leveraging a feature pyramid with cross-attentive pooling, our approach effectively manages the trade-off between information loss and efficiency without being affected by varying document image sizes. Furthermore, we introduce a novel instruction tuning task, which facilitates the model's text-reading capability by learning to predict the relative positions of input text, eventually minimizing the risk of truncated text caused by the limited capacity of LLMs. Comprehensive experiments validate the effectiveness of our approach, demonstrating superior performance in various document understanding tasks.

Open-vocabulary semantic segmentation (OVSS) aims to segment unseen classes without corresponding labels. Existing Vision-Language Model (VLM)-based methods leverage VLM's rich knowledge to enhance additional explicit segmentation-specific networks, yielding competitive results, but at the cost of extensive training cost. To reduce the cost, we attempt to enable VLM to directly produce the segmentation results without any segmentation-specific networks. Prompt learning offers a direct and parameter-efficient approach, yet it falls short in guiding VLM for pixel-level visual classification. Therefore, we propose the ${\bf R}$elationship ${\bf P}$rompt ${\bf M}$odule (${\bf RPM}$), which generates the relationship prompt that directs VLM to extract pixel-level semantic embeddings suitable for OVSS. Moreover, RPM integrates with VLM to construct the ${\bf R}$elationship ${\bf P}$rompt ${\bf N}$etwork (${\bf RPN}$), achieving OVSS without any segmentation-specific networks. RPN attains state-of-the-art performance with merely about ${\bf 3M}$ trainable parameters (2\% of total parameters).

The ability of large language models (LLMs) to mimic human-like intelligence has led to a surge in LLM-based autonomous agents. Though recent LLMs seem capable of planning and reasoning given user instructions, their effectiveness in applying these capabilities for autonomous task solving remains underexplored. This is especially true in enterprise settings, where automated agents hold the promise of a high impact. To fill this gap, we propose WorkArena++, a novel benchmark consisting of 682 tasks corresponding to realistic workflows routinely performed by knowledge workers. WorkArena++ is designed to evaluate the planning, problem-solving, logical/arithmetic reasoning, retrieval, and contextual understanding abilities of web agents. Our empirical studies across state-of-the-art LLMs and vision-language models (VLMs), as well as human workers, reveal several challenges for such models to serve as useful assistants in the workplace. In addition to the benchmark, we provide a mechanism to effortlessly generate thousands of ground-truth observation/action traces, which can be used for fine-tuning existing models. Overall, we expect this work to serve as a useful resource to help the community progress towards capable autonomous agents. The benchmark can be found at https://github.com/ServiceNow/WorkArena.

We introduce FlexCap, a vision-language model that generates region-specific descriptions of varying lengths. FlexCap is trained to produce length-conditioned captions for input boxes, enabling control over information density, with descriptions ranging from concise object labels to detailed captions. To achieve this, we create large-scale training datasets of image region descriptions with varying lengths from captioned web images. We demonstrate FlexCap’s effectiveness in several applications: first, it achieves strong performance in dense captioning tasks on the Visual Genome dataset. Second, we show how FlexCap’s localized descriptions can serve as input to a large language model to create a visual question answering (VQA) system, achieving state-of-the-art zero-shot performance on multiple VQA benchmarks. Our experiments illustrate FlexCap’s utility for tasks including image labeling, object attribute recognition, and visual dialog. Project webpage: https://flex-cap.github.io.

A frequent problem in vision-based reasoning tasks such as object detection and optical character recognition (OCR) is the persistence of specular highlights. Specular highlights appear as bright spots of glare that occur due to the concentrated reflection of light; these spots manifest as image artifacts which occlude computer vision models and are challenging to reconstruct. Despite this, specular highlight removal receives relatively little attention due to the difficulty of acquiring high-quality, real-world data. We introduce a method to generate specular highlight data with near-perfect alignment and present SHDocs—a dataset of specular highlights on document images created using our method. Through our benchmark, we demonstrate that our dataset enables us to surpass the performance of state-of-the-art specular highlight removal models and downstream OCR tasks. We release our dataset, code, and methods publicly to motivate further exploration of image enhancement for practical computer vision challenges.

Compositional Reasoning (CR) entails grasping the significance of attributes, relations, and word order. Recent Vision-Language Models (VLMs), comprising a visual encoder and a Large Language Model (LLM) decoder, have demonstrated remarkable proficiency in such reasoning tasks. This prompts a crucial question: have VLMs effectively tackled the CR challenge? We conjecture that existing CR benchmarks may not adequately push the boundaries of modern VLMs due to the reliance on an LLM only negative text generation pipeline. Consequently, the negatives produced either appear as outliers from the natural language distribution learned by VLMs' LLM decoders or as improbable within the corresponding image context. To address these limitations, we introduce ConMe\footnote{ConMe is an abbreviation for Confuse Me.} -- a compositional reasoning benchmark and a novel data generation pipeline leveraging VLMs to produce `hard CR Q&A'. Through a new concept of VLMs conversing with each other to collaboratively expose their weaknesses, our pipeline autonomously generates, evaluates, and selects challenging compositional reasoning questions, establishing a robust CR benchmark, also subsequently validated manually. Our benchmark provokes a noteworthy, up to 33%, decrease in CR performance compared to preceding benchmarks, reinstating the CR challenge even for state-of-the-art VLMs.