arXiv subjects

Junnan Li

Publications and source records attributed to Junnan Li.

At least 55 records · Page 3Linked to original sources

arXiv2024

Accelerating Video Diffusion Models via Distribution Matching

Generative models, particularly diffusion models, have made significant success in data synthesis across various modalities, including images, videos, and 3D assets. However, current diffusion models are computationally intensive, often requiring numerous sampling steps that limit their practical application, especially in video generation. This work introduces a novel framework for diffusion distillation and distribution matching that dramatically reduces the number of inference steps while maintaining-and potentially improving-generation quality. Our approach focuses on distilling pre-trained diffusion models into a more efficient few-step generator, specifically targeting video generation. By leveraging a combination of video GAN loss and a novel 2D score distribution matching loss, we demonstrate the potential to generate high-quality video frames with substantially fewer sampling steps. To be specific, the proposed method incorporates a denoising GAN discriminator to distil from the real data and a pre-trained image diffusion model to enhance the frame quality and the prompt-following capabilities. Experimental results using AnimateDiff as the teacher model showcase the method's effectiveness, achieving superior performance in just four sampling steps compared to existing techniques.

cs.CV↗

arXiv2024

Object Augmentation Algorithm: Computing virtual object motion and object induced interaction wrench from optical markers

This study addresses the critical need for diverse and comprehensive data focused on human arm joint torques while performing activities of daily living (ADL). Previous studies have often overlooked the influence of objects on joint torques during ADL, resulting in limited datasets for analysis. To address this gap, we propose an Object Augmentation Algorithm (OAA) capable of augmenting existing marker-based databases with virtual object motions and object-induced joint torque estimations. The OAA consists of five phases: (1) computing hand coordinate systems from optical markers, (2) characterising object movements with virtual markers, (3) calculating object motions through inverse kinematics (IK), (4) determining the wrench necessary for prescribed object motion using inverse dynamics (ID), and (5) computing joint torques resulting from object manipulation. The algorithm's accuracy is validated through trajectory tracking and torque analysis on a 5+4 degree of freedom (DoF) robotic hand-arm system, manipulating three unique objects. The results show that the OAA can accurately and precisely estimate 6 DoF object motion and object-induced joint torques. Correlations between computed and measured quantities were > 0.99 for object trajectories and > 0.93 for joint torques. The OAA was further shown to be robust to variations in the number and placement of input markers, which are expected between databases. Differences between repeated experiments were minor but significant (p < 0.05). The algorithm expands the scope of available data and facilitates more comprehensive analyses of human-object interaction dynamics.

cs.RO↗

arXiv2024

X-InstructBLIP: A Framework for aligning X-Modal instruction-aware representations to LLMs and Emergent Cross-modal Reasoning

Recent research has achieved significant advancements in visual reasoning tasks through learning image-to-language projections and leveraging the impressive reasoning abilities of Large Language Models (LLMs). This paper introduces an efficient and effective framework that integrates multiple modalities (images, 3D, audio and video) to a frozen LLM and demonstrates an emergent ability for cross-modal reasoning (2+ modality inputs). Our approach explores two distinct projection mechanisms: Q-Formers and Linear Projections (LPs). Through extensive experimentation across all four modalities on 16 benchmarks, we explore both methods and assess their adaptability in integrated and separate cross-modal reasoning. The Q-Former projection demonstrates superior performance in single modality scenarios and adaptability in joint versus discriminative reasoning involving two or more modalities. However, it exhibits lower generalization capabilities than linear projection in contexts where task-modality data are limited. To enable this framework, we devise a scalable pipeline that automatically generates high-quality, instruction-tuning datasets from readily available captioning data across different modalities, and contribute 24K QA data for audio and 250K QA data for 3D. To facilitate further research in cross-modal reasoning, we introduce the DisCRn (Discriminative Cross-modal Reasoning) benchmark comprising 9K audio-video QA samples and 28K image-3D QA samples that require the model to reason discriminatively across disparate input modalities.

cs.CV↗

arXiv2024

Optimizing Interaction Space: Enlarging the Capture Volume for Multiple Portable Motion Capture Devices

Markerless motion capture devices such as the Leap Motion Controller (LMC) have been extensively used for tracking hand, wrist, and forearm positions as an alternative to Marker-based Motion Capture (MMC). However, previous studies have highlighted the subpar performance of LMC in reliably recording hand kinematics. In this study, we employ four LMC devices to optimize their collective tracking volume, aiming to enhance the accuracy and precision of hand kinematics. Through Monte Carlo simulation, we determine an optimized layout for the four LMC devices and subsequently conduct reliability and validity experiments encompassing 1560 trials across ten subjects. The combined tracking volume is validated against an MMC system, particularly for kinematic movements involving wrist, index, and thumb flexion. Utilizing calculation resources in one computer, our result of the optimized configuration has a better visibility rate with a value of 0.05 $\pm$ 0.55 compared to the initial configuration with -0.07 $\pm$ 0.40. Multiple Leap Motion Controllers (LMCs) have proven to increase the interaction space of capture volume but are still unable to give agreeable measurements from dynamic movement.

cs.RO↗

arXiv2024

Identification and validation of the dynamic model of a tendon-driven anthropomorphic finger

This study addresses the absence of an identification framework to quantify a comprehensive dynamic model of human and anthropomorphic tendon-driven fingers, which is necessary to investigate the physiological properties of human fingers and improve the control of robotic hands. First, a generalized dynamic model was formulated, which takes into account the inherent properties of such a mechanical system. This includes rigid-body dynamics, coupling matrix, joint viscoelasticity, and tendon friction. Then, we propose a methodology comprising a series of experiments, for step-wise identification and validation of this dynamic model. Moreover, an experimental setup was designed and constructed that features actuation modules and peripheral sensors to facilitate the identification process. To verify the proposed methodology, a 3D-printed robotic finger based on the index finger design of the Dexmart hand was developed, and the proposed experiments were executed to identify and validate its dynamic model. This study could be extended to explore the identification of cadaver hands, aiming for a consistent dataset from a single cadaver specimen to improve the development of musculoskeletal hand models.

cs.RO↗

arXiv2024

LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding

Large multimodal models (LMMs) are processing increasingly longer and richer inputs. Albeit the progress, few public benchmark is available to measure such development. To mitigate this gap, we introduce LongVideoBench, a question-answering benchmark that features video-language interleaved inputs up to an hour long. Our benchmark includes 3,763 varying-length web-collected videos with their subtitles across diverse themes, designed to comprehensively evaluate LMMs on long-term multimodal understanding. To achieve this, we interpret the primary challenge as to accurately retrieve and reason over detailed multimodal information from long inputs. As such, we formulate a novel video question-answering task termed referring reasoning. Specifically, as part of the question, it contains a referring query that references related video contexts, called referred context. The model is then required to reason over relevant video details from the referred context. Following the paradigm of referring reasoning, we curate 6,678 human-annotated multiple-choice questions in 17 fine-grained categories, establishing one of the most comprehensive benchmarks for long-form video understanding. Evaluations suggest that the LongVideoBench presents significant challenges even for the most advanced proprietary models (e.g. GPT-4o, Gemini-1.5-Pro, GPT-4-Turbo), while their open-source counterparts show an even larger performance gap. In addition, our results indicate that model performance on the benchmark improves only when they are capable of processing more frames, positioning LongVideoBench as a valuable benchmark for evaluating future-generation long-context LMMs.

cs.CV↗

arXiv2024

ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding

Recent advancements in multimodal pre-training have shown promising efficacy in 3D representation learning by aligning multimodal features across 3D shapes, their 2D counterparts, and language descriptions. However, the methods used by existing frameworks to curate such multimodal data, in particular language descriptions for 3D shapes, are not scalable, and the collected language descriptions are not diverse. To address this, we introduce ULIP-2, a simple yet effective tri-modal pre-training framework that leverages large multimodal models to automatically generate holistic language descriptions for 3D shapes. It only needs 3D data as input, eliminating the need for any manual 3D annotations, and is therefore scalable to large datasets. ULIP-2 is also equipped with scaled-up backbones for better multimodal representation learning. We conduct experiments on two large-scale 3D datasets, Objaverse and ShapeNet, and augment them with tri-modal datasets of 3D point clouds, images, and language for training ULIP-2. Experiments show that ULIP-2 demonstrates substantial benefits in three downstream tasks: zero-shot 3D classification, standard 3D classification with fine-tuning, and 3D captioning (3D-to-language generation). It achieves a new SOTA of 50.6% (top-1) on Objaverse-LVIS and 84.7% (top-1) on ModelNet40 in zero-shot classification. In the ScanObjectNN benchmark for standard fine-tuning, ULIP-2 reaches an overall accuracy of 91.5% with a compact model of only 1.4 million parameters. ULIP-2 sheds light on a new paradigm for scalable multimodal 3D representation learning without human annotations and shows significant improvements over existing baselines. The code and datasets are released at https://github.com/salesforce/ULIP.

cs.CV↗

arXiv2024

What Are We Measuring When We Evaluate Large Vision-Language Models? An Analysis of Latent Factors and Biases

Vision-language (VL) models, pretrained on colossal image-text datasets, have attained broad VL competence that is difficult to evaluate. A common belief is that a small number of VL skills underlie the variety of VL tests. In this paper, we perform a large-scale transfer learning experiment aimed at discovering latent VL skills from data. We reveal interesting characteristics that have important implications for test suite design. First, generation tasks suffer from a length bias, suggesting benchmarks should balance tasks with varying output lengths. Second, we demonstrate that factor analysis successfully identifies reasonable yet surprising VL skill factors, suggesting benchmarks could leverage similar analyses for task selection. Finally, we present a new dataset, OLIVE (https://github.com/jq-zh/olive-dataset), which simulates user instructions in the wild and presents challenges dissimilar to all datasets we tested. Our findings contribute to the design of balanced and broad-coverage vision-language evaluation methods.

cs.CV↗

arXiv2023

Tackling Data Heterogeneity in Federated Learning with Class Prototypes

Data heterogeneity across clients in federated learning (FL) settings is a widely acknowledged challenge. In response, personalized federated learning (PFL) emerged as a framework to curate local models for clients' tasks. In PFL, a common strategy is to develop local and global models jointly - the global model (for generalization) informs the local models, and the local models (for personalization) are aggregated to update the global model. A key observation is that if we can improve the generalization ability of local models, then we can improve the generalization of global models, which in turn builds better personalized models. In this work, we consider class imbalance, an overlooked type of data heterogeneity, in the classification setting. We propose FedNH, a novel method that improves the local models' performance for both personalization and generalization by combining the uniformity and semantics of class prototypes. FedNH initially distributes class prototypes uniformly in the latent space and smoothly infuses the class semantics into class prototypes. We show that imposing uniformity helps to combat prototype collapse while infusing class semantics improves local models. Extensive experiments were conducted on popular classification datasets under the cross-device setting. Our results demonstrate the effectiveness and stability of our method over recent works.

cs.LG↗

arXiv2023

BLIP-Diffusion: Pre-trained Subject Representation for Controllable Text-to-Image Generation and Editing

Subject-driven text-to-image generation models create novel renditions of an input subject based on text prompts. Existing models suffer from lengthy fine-tuning and difficulties preserving the subject fidelity. To overcome these limitations, we introduce BLIP-Diffusion, a new subject-driven image generation model that supports multimodal control which consumes inputs of subject images and text prompts. Unlike other subject-driven generation models, BLIP-Diffusion introduces a new multimodal encoder which is pre-trained to provide subject representation. We first pre-train the multimodal encoder following BLIP-2 to produce visual representation aligned with the text. Then we design a subject representation learning task which enables a diffusion model to leverage such visual representation and generates new subject renditions. Compared with previous methods such as DreamBooth, our model enables zero-shot subject-driven generation, and efficient fine-tuning for customized subject with up to 20x speedup. We also demonstrate that BLIP-Diffusion can be flexibly combined with existing techniques such as ControlNet and prompt-to-prompt to enable novel subject-driven generation and editing applications. Code and models will be released at https://github.com/salesforce/LAVIS/tree/main/projects/blip-diffusion. Project page at https://dxli94.github.io/BLIP-Diffusion-website/.

cs.CV↗

arXiv2023

BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models

The cost of vision-and-language pre-training has become increasingly prohibitive due to end-to-end training of large-scale models. This paper proposes BLIP-2, a generic and efficient pre-training strategy that bootstraps vision-language pre-training from off-the-shelf frozen pre-trained image encoders and frozen large language models. BLIP-2 bridges the modality gap with a lightweight Querying Transformer, which is pre-trained in two stages. The first stage bootstraps vision-language representation learning from a frozen image encoder. The second stage bootstraps vision-to-language generative learning from a frozen language model. BLIP-2 achieves state-of-the-art performance on various vision-language tasks, despite having significantly fewer trainable parameters than existing methods. For example, our model outperforms Flamingo80B by 8.7% on zero-shot VQAv2 with 54x fewer trainable parameters. We also demonstrate the model's emerging capabilities of zero-shot image-to-text generation that can follow natural language instructions.

cs.CV↗

arXiv2023

InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning

Large-scale pre-training and instruction tuning have been successful at creating general-purpose language models with broad competence. However, building general-purpose vision-language models is challenging due to the rich input distributions and task diversity resulting from the additional visual input. Although vision-language pretraining has been widely studied, vision-language instruction tuning remains under-explored. In this paper, we conduct a systematic and comprehensive study on vision-language instruction tuning based on the pretrained BLIP-2 models. We gather 26 publicly available datasets, covering a wide variety of tasks and capabilities, and transform them into instruction tuning format. Additionally, we introduce an instruction-aware Query Transformer, which extracts informative features tailored to the given instruction. Trained on 13 held-in datasets, InstructBLIP attains state-of-the-art zero-shot performance across all 13 held-out datasets, substantially outperforming BLIP-2 and larger Flamingo models. Our models also lead to state-of-the-art performance when finetuned on individual downstream tasks (e.g., 90.7% accuracy on ScienceQA questions with image contexts). Furthermore, we qualitatively demonstrate the advantages of InstructBLIP over concurrent multimodal models. All InstructBLIP models are open-sourced at https://github.com/salesforce/LAVIS/tree/main/projects/instructblip.

cs.CV↗

arXiv2023

CodeT5+: Open Code Large Language Models for Code Understanding and Generation

Large language models (LLMs) pretrained on vast source code have achieved prominent progress in code intelligence. However, existing code LLMs have two main limitations in terms of architecture and pretraining tasks. First, they often adopt a specific architecture (encoder-only or decoder-only) or rely on a unified encoder-decoder network for different downstream tasks. The former paradigm is limited by inflexibility in applications while in the latter, the model is treated as a single system for all tasks, leading to suboptimal performance on a subset of tasks. Secondly, they often employ a limited set of pretraining objectives which might not be relevant to some downstream tasks and hence result in substantial performance degrade. To address these limitations, we propose ``CodeT5+'', a family of encoder-decoder LLMs for code in which component modules can be flexibly combined to suit a wide range of downstream code tasks. Such flexibility is enabled by our proposed mixture of pretraining objectives to mitigate the pretrain-finetune discrepancy. These objectives cover span denoising, contrastive learning, text-code matching, and causal LM pretraining tasks, on both unimodal and bimodal multilingual code corpora. Furthermore, we propose to initialize CodeT5+ with frozen off-the-shelf LLMs without training from scratch to efficiently scale up our models, and explore instruction-tuning to align with natural language instructions. We extensively evaluate CodeT5+ on over 20 code-related benchmarks in different settings, including zero-shot, finetuning, and instruction-tuning. We observe state-of-the-art (SoTA) model performance on various code-related tasks, such as code generation and completion, math programming, and text-to-code retrieval tasks. Particularly, our instruction-tuned CodeT5+ 16B achieves new SoTA results on HumanEval code generation task against other open code LLMs.

cs.CL↗

arXiv2023

From Images to Textual Prompts: Zero-shot VQA with Frozen Large Language Models

Large language models (LLMs) have demonstrated excellent zero-shot generalization to new language tasks. However, effective utilization of LLMs for zero-shot visual question-answering (VQA) remains challenging, primarily due to the modality disconnection and task disconnection between LLM and VQA task. End-to-end training on vision and language data may bridge the disconnections, but is inflexible and computationally expensive. To address this issue, we propose \emph{Img2Prompt}, a plug-and-play module that provides the prompts that can bridge the aforementioned modality and task disconnections, so that LLMs can perform zero-shot VQA tasks without end-to-end training. In order to provide such prompts, we further employ LLM-agnostic models to provide prompts that can describe image content and self-constructed question-answer pairs, which can effectively guide LLM to perform zero-shot VQA tasks. Img2Prompt offers the following benefits: 1) It can flexibly work with various LLMs to perform VQA. 2)~Without the needing of end-to-end training, it significantly reduces the cost of deploying LLM for zero-shot VQA tasks. 3) It achieves comparable or better performance than methods relying on end-to-end training. For example, we outperform Flamingo \cite{Deepmind:Flamingo2022} by 5.6\% on VQAv2. On the challenging A-OKVQA dataset, our method even outperforms few-shot methods by as much as 20\%.

cs.CV↗

arXiv2023

Improving Tail-Class Representation with Centroid Contrastive Learning

In vision domain, large-scale natural datasets typically exhibit long-tailed distribution which has large class imbalance between head and tail classes. This distribution poses difficulty in learning good representations for tail classes. Recent developments have shown good long-tailed model can be learnt by decoupling the training into representation learning and classifier balancing. However, these works pay insufficient consideration on the long-tailed effect on representation learning. In this work, we propose interpolative centroid contrastive learning (ICCL) to improve long-tailed representation learning. ICCL interpolates two images from a class-agnostic sampler and a class-aware sampler, and trains the model such that the representation of the interpolative image can be used to retrieve the centroids for both source classes. We demonstrate the effectiveness of our approach on multiple long-tailed image classification benchmarks. Our result shows a significant accuracy gain of 2.8% on the iNaturalist 2018 dataset with a real-world long-tailed distribution.

cs.CV↗

arXiv2023

Plug-and-Play VQA: Zero-shot VQA by Conjoining Large Pretrained Models with Zero Training

Visual question answering (VQA) is a hallmark of vision and language reasoning and a challenging task under the zero-shot setting. We propose Plug-and-Play VQA (PNP-VQA), a modular framework for zero-shot VQA. In contrast to most existing works, which require substantial adaptation of pretrained language models (PLMs) for the vision modality, PNP-VQA requires no additional training of the PLMs. Instead, we propose to use natural language and network interpretation as an intermediate representation that glues pretrained models together. We first generate question-guided informative image captions, and pass the captions to a PLM as context for question answering. Surpassing end-to-end trained baselines, PNP-VQA achieves state-of-the-art results on zero-shot VQAv2 and GQA. With 11B parameters, it outperforms the 80B-parameter Flamingo model by 8.5% on VQAv2. With 738M PLM parameters, PNP-VQA achieves an improvement of 9.1% on GQA over FewVLM with 740M PLM parameters. Code is released at https://github.com/salesforce/LAVIS/tree/main/projects/pnp-vqa

cs.CV↗

arXiv2023

Masked Unsupervised Self-training for Label-free Image Classification

State-of-the-art computer vision models are mostly trained with supervised learning using human-labeled images, which limits their scalability due to the expensive annotation cost. While self-supervised representation learning has achieved impressive progress, it still requires a second stage of finetuning on labeled data. On the other hand, models pre-trained with large-scale text-image supervision (e.g., CLIP) have enabled zero-shot transfer to downstream image classification tasks. However, the zero-shot performance of CLIP-like models are often insufficient for real-world adoption. In this paper, we aim to leverage the abundant unlabeled data from a target domain to improve the performance of a pre-trained zero-shot classifier, by unsupervised finetuning of the pre-trained model. We propose Masked Unsupervised Self-Training (MUST), a new unsupervised adaptation method which leverages two different and complementary sources of training signals: pseudo-labels and raw images. MUST jointly optimizes three objectives to learn both class-level global feature and pixel-level local feature and enforces a regularization between the two. We demonstrate the efficacy of MUST on a variety of downstream tasks, where it improves upon CLIP by a large margin. MUST also outperforms supervised few-shot adaptation methods. It achieves a top-1 accuracy of 77.7% on ImageNet using ViT-B, +9.4% higher than CLIP, and +6.2% higher than 16-shot CLIP adaptation. Our code is available at https://github.com/salesforce/MUST.

cs.CV↗

arXiv2022

BotSIM: An End-to-End Bot Simulation Toolkit for Commercial Task-Oriented Dialog Systems

We introduce BotSIM, a modular, open-source Bot SIMulation environment with dialog generation, user simulation and conversation analytics capabilities. BotSIM aims to serve as a one-stop solution for large-scale data-efficient end-to-end evaluation, diagnosis and remediation of commercial task-oriented dialog (TOD) systems to significantly accelerate commercial bot development and evaluation, reduce cost and time-to-market. BotSIM adopts a layered design comprising the infrastructure layer, the adaptor layer and the application layer. The infrastructure layer hosts key models and components to support BotSIM's major functionalities via a streamlined "generation-simulation-remediation" pipeline. The adaptor layer is used to extend BotSIM to accommodate new bot platforms. The application layer provides a suite of command line tools and a Web App to significantly lower the entry barrier for BotSIM users such as bot admins or practitioners. In this report, we focus on the technical designs of various system components. A detailed case study using Einstein BotBuilder is also presented to show how to apply BotSIM pipeline for bot evaluation and remediation. The detailed system descriptions can be found in our system demo paper. The toolkit is available at: https://github.com/salesforce/BotSIM .

cs.CL↗