InstructDET: Diversifying Referring Object Detection with Generalized Instructions

We propose InstructDET, a data-centric method for referring object detection (ROD) that localizes target objects based on user instructions. While deriving from referring expressions (REC), the instructions we leverage are greatly diversified to encompass common user intentions related to object detection. For one image, we produce tremendous instructions that refer to every single object and different combinations of multiple objects. Each instruction and its corresponding object bounding boxes (bbxs) constitute one training data pair. In order to encompass common detection expressions, we involve emerging vision-language model (VLM) and large language model (LLM) to generate instructions guided by text prompts and object bbxs, as the generalizations of foundation models are effective to produce human-like expressions (e.g., describing object property, category, and relationship). We name our constructed dataset as InDET. It contains images, bbxs and generalized instructions that are from foundation models. Our InDET is developed from existing REC datasets and object detection datasets, with the expanding potential that any image with object bbxs can be incorporated through using our InstructDET method. By using our InDET dataset, we show that a conventional ROD model surpasses existing methods on standard REC datasets and our InDET test set. Our data-centric method InstructDET, with automatic data expansion by leveraging foundation models, directs a promising field that ROD can be greatly diversified to execute common object detection instructions.

DRAG: Divergence-based Adaptive Aggregation in Federated learning on Non-IID Data

Local stochastic gradient descent (SGD) is a fundamental approach in achieving communication efficiency in Federated Learning (FL) by allowing individual workers to perform local updates. However, the presence of heterogeneous data distributions across working nodes causes each worker to update its local model towards a local optimum, leading to the phenomenon known as ``client-drift" and resulting in slowed convergence. To address this issue, previous works have explored methods that either introduce communication overhead or suffer from unsteady performance. In this work, we introduce a novel metric called ``degree of divergence," quantifying the angle between the local gradient and the global reference direction. Leveraging this metric, we propose the divergence-based adaptive aggregation (DRAG) algorithm, which dynamically ``drags" the received local updates toward the reference direction in each round without requiring extra communication overhead. Furthermore, we establish a rigorous convergence analysis for DRAG, proving its ability to achieve a sublinear convergence rate. Compelling experimental results are presented to illustrate DRAG's superior performance compared to state-of-the-art algorithms in effectively managing the client-drift phenomenon. Additionally, DRAG exhibits remarkable resilience against certain Byzantine attacks. By securely sharing a small sample of the client's data with the FL server, DRAG effectively counters these attacks, as demonstrated through comprehensive experiments.

Freshness or Accuracy, Why Not Both? Addressing Delayed Feedback via Dynamic Graph Neural Networks

The delayed feedback problem is one of the most pressing challenges in predicting the conversion rate since users' conversions are always delayed in online commercial systems. Although new data are beneficial for continuous training, without complete feedback information, i.e., conversion labels, training algorithms may suffer from overwhelming fake negatives. Existing methods tend to use multitask learning or design data pipelines to solve the delayed feedback problem. However, these methods have a trade-off between data freshness and label accuracy. In this paper, we propose Delayed Feedback Modeling by Dynamic Graph Neural Network (DGDFEM). It includes three stages, i.e., preparing a data pipeline, building a dynamic graph, and training a CVR prediction model. In the model training, we propose a novel graph convolutional method named HLGCN, which leverages both high-pass and low-pass filters to deal with conversion and non-conversion relationships. The proposed method achieves both data freshness and label accuracy. We conduct extensive experiments on three industry datasets, which validate the consistent superiority of our method.

Link-Context Learning for Multimodal LLMs

The ability to learn from context with novel concepts, and deliver appropriate responses are essential in human conversations. Despite current Multimodal Large Language Models (MLLMs) and Large Language Models (LLMs) being trained on mega-scale datasets, recognizing unseen images or understanding novel concepts in a training-free manner remains a challenge. In-Context Learning (ICL) explores training-free few-shot learning, where models are encouraged to ``learn to learn" from limited tasks and generalize to unseen tasks. In this work, we propose link-context learning (LCL), which emphasizes "reasoning from cause and effect" to augment the learning capabilities of MLLMs. LCL goes beyond traditional ICL by explicitly strengthening the causal relationship between the support set and the query set. By providing demonstrations with causal links, LCL guides the model to discern not only the analogy but also the underlying causal associations between data points, which empowers MLLMs to recognize unseen images and understand novel concepts more effectively. To facilitate the evaluation of this novel approach, we introduce the ISEKAI dataset, comprising exclusively of unseen generated image-label pairs designed for link-context learning. Extensive experiments show that our LCL-MLLM exhibits strong link-context learning capabilities to novel concepts over vanilla MLLMs. Code and data will be released at https://github.com/isekai-portal/Link-Context-Learning.

Relation-Aware Distribution Representation Network for Person Clustering with Multiple Modalities

Person clustering with multi-modal clues, including faces, bodies, and voices, is critical for various tasks, such as movie parsing and identity-based movie editing. Related methods such as multi-view clustering mainly project multi-modal features into a joint feature space. However, multi-modal clue features are usually rather weakly correlated due to the semantic gap from the modality-specific uniqueness. As a result, these methods are not suitable for person clustering. In this paper, we propose a Relation-Aware Distribution representation Network (RAD-Net) to generate a distribution representation for multi-modal clues. The distribution representation of a clue is a vector consisting of the relation between this clue and all other clues from all modalities, thus being modality agnostic and good for person clustering. Accordingly, we introduce a graph-based method to construct distribution representation and employ a cyclic update policy to refine distribution representation progressively. Our method achieves substantial improvements of +6% and +8.2% in F-score on the Video Person-Clustering Dataset (VPCD) and VoxCeleb2 multi-view clustering dataset, respectively. Codes will be released publicly upon acceptance.

Domain Disentanglement with Interpolative Data Augmentation for Dual-Target Cross-Domain Recommendation

The conventional single-target Cross-Domain Recommendation (CDR) aims to improve the recommendation performance on a sparser target domain by transferring the knowledge from a source domain that contains relatively richer information. By contrast, in recent years, dual-target CDR has been proposed to improve the recommendation performance on both domains simultaneously. However, to this end, there are two challenges in dual-target CDR: (1) how to generate both relevant and diverse augmented user representations, and (2) how to effectively decouple domain-independent information from domain-specific information, in addition to domain-shared information, to capture comprehensive user preferences. To address the above two challenges, we propose a Disentanglement-based framework with Interpolative Data Augmentation for dual-target Cross-Domain Recommendation, called DIDA-CDR. In DIDA-CDR, we first propose an interpolative data augmentation approach to generating both relevant and diverse augmented user representations to augment sparser domain and explore potential user preferences. We then propose a disentanglement module to effectively decouple domain-specific and domain-independent information to capture comprehensive user preferences. Both steps significantly contribute to capturing more comprehensive user preferences, thereby improving the recommendation performance on each domain. Extensive experiments conducted on five real-world datasets show the significant superiority of DIDA-CDR over the state-of-the-art methods.

Exposing the Troublemakers in Described Object Detection

Detecting objects based on language descriptions is a popular task that includes Open-Vocabulary object Detection (OVD) and Referring Expression Comprehension (REC). In this paper, we advance them to a more practical setting called Described Object Detection (DOD) by expanding category names to flexible language expressions for OVD and overcoming the limitation of REC to only grounding the pre-existing object. We establish the research foundation for DOD tasks by constructing a Description Detection Dataset ($D^3$), featuring flexible language expressions and annotating all described objects without omission. By evaluating previous SOTA methods on $D^3$, we find some troublemakers that fail current REC, OVD, and bi-functional methods. REC methods struggle with confidence scores, rejecting negative instances, and multi-target scenarios, while OVD methods face constraints with long and complex descriptions. Recent bi-functional methods also do not work well on DOD due to their separated training procedures and inference strategies for REC and OVD tasks. Building upon the aforementioned findings, we propose a baseline that largely improves REC methods by reconstructing the training data and introducing a binary classification sub-task, outperforming existing methods. Data and code is available at https://github.com/shikras/d-cube.

Retrieve Anyone: A General-purpose Person Re-identification Task with Instructions

Human intelligence can retrieve any person according to both visual and language descriptions. However, the current computer vision community studies specific person re-identification (ReID) tasks in different scenarios separately, which limits the applications in the real world. This paper strives to resolve this problem by proposing a new instruct-ReID task that requires the model to retrieve images according to the given image or language instructions.Our instruct-ReID is a more general ReID setting, where existing ReID tasks can be viewed as special cases by designing different instructions. We propose a large-scale OmniReID benchmark and an adaptive triplet loss as a baseline method to facilitate research in this new setting. Experimental results show that the baseline model trained on our OmniReID benchmark can improve +0.6%, +1.4%, 0.2% mAP on Market1501, CUHK03, MSMT17 for traditional ReID, +0.8%, +2.0%, +13.4% mAP on PRCC, VC-Clothes, LTCC for clothes-changing ReID, +11.7% mAP on COCAS+ real2 for clothestemplate based clothes-changing ReID when using only RGB images, +25.4% mAP on COCAS+ real2 for our newly defined language-instructed ReID. The dataset, model, and code will be available at https://github.com/hwz-zju/Instruct-ReID.

Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic

In human conversations, individuals can indicate relevant regions within a scene while addressing others. In turn, the other person can then respond by referring to specific regions if necessary. This natural referential ability in dialogue remains absent in current Multimodal Large Language Models (MLLMs). To fill this gap, this paper proposes an MLLM called Shikra, which can handle spatial coordinate inputs and outputs in natural language. Its architecture consists of a vision encoder, an alignment layer, and a LLM. It is designed to be straightforward and simple, without the need for extra vocabularies, position encoder, pre-/post-detection modules, or external plug-in models. All inputs and outputs are in natural language form. Referential dialogue is a superset of various vision-language (VL) tasks. Shikra can naturally handle location-related tasks like REC and PointQA, as well as conventional VL tasks such as Image Captioning and VQA. Experimental results showcase Shikra's promising performance. Furthermore, it enables numerous exciting applications, like providing mentioned objects' coordinates in chains of thoughts and comparing user-pointed regions similarities. Our code, model and dataset are accessed at https://github.com/shikras/shikra.