Vision-and-Language Models (VLMs) have shown impressive capabilities on single-turn benchmarks, yet real-world applications often demand more intricate multi-turn dialogues. Existing multi-turn datasets (e.g, MMDU, ConvBench) only partially capture the breadth and depth of conversational scenarios encountered by users. In this work, we introduce MultiVerse, a novel multi-turn conversation benchmark featuring 647 dialogues - each averaging four turns - derived from a diverse set of 12 popular VLM evaluation benchmarks. With 484 tasks and 484 interaction goals, MultiVerse covers a wide range of topics, from factual knowledge and perception to advanced reasoning tasks such as mathematics and coding. To facilitate robust assessment, we propose a checklist-based evaluation method that leverages GPT-4o as the automated evaluator, measuring performance across 37 key aspects, including perceptual accuracy, linguistic clarity, and factual correctness. We evaluate 18 VLMs on MultiVerse, revealing that even the strongest models (e.g., GPT-4o) achieve only a 50% success rate in complex multi-turn conversations, highlighting the dataset's challenging nature. Notably, we find that providing full dialogue context significantly enhances performance for smaller or weaker models, emphasizing the importance of in-context learning. We believe MultiVerse is a landscape of evaluating multi-turn interaction abilities for VLMs.

In content-based video retrieval (CBVR), dealing with large-scale collections, efficiency is as important as accuracy; thus, several video-level feature-based studies have actively been conducted. Nevertheless, owing to the severe difficulty of embedding a lengthy and untrimmed video into a single feature, these studies have been insufficient for accurate retrieval compared to frame-level feature-based studies. In this paper, we show that appropriate suppression of irrelevant frames can provide insight into the current obstacles of the video-level approaches. Furthermore, we propose a Video-to-Video Suppression network (VVS) as a solution. VVS is an end-to-end framework that consists of an easy distractor elimination stage to identify which frames to remove and a suppression weight generation stage to determine the extent to suppress the remaining frames. This structure is intended to effectively describe an untrimmed video with varying content and meaningless information. Its efficacy is proved via extensive experiments, and we show that our approach is not only state-of-the-art in video-level approaches but also has a fast inference time despite possessing retrieval capabilities close to those of frame-level approaches. Code is available at https://github.com/sejong-rcv/VVS

Young-Jun Lee, Byungsoo Ko, Han-Gyu Kim, Jonghwan Hyeon, Ho-Jin Choi. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Previous deep learning-based line segment detection (LSD) suffers from the immense model size and high computational cost for line prediction. This constrains them from real-time inference on computationally restricted environments. In this paper, we propose a real-time and light-weight line segment detector for resource-constrained environments named Mobile LSD (M-LSD). We design an extremely efficient LSD architecture by minimizing the backbone network and removing the typical multi-module process for line prediction found in previous methods. To maintain competitive performance with a light-weight network, we present novel training schemes: Segments of Line segment (SoL) augmentation, matching and geometric loss. SoL augmentation splits a line segment into multiple subparts, which are used to provide auxiliary line data during the training process. Moreover, the matching and geometric loss allow a model to capture additional geometric cues. Compared with TP-LSD-Lite, previously the best real-time LSD method, our model (M-LSD-tiny) achieves competitive performance with 2.5% of model size and an increase of 130.5% in inference speed on GPU. Furthermore, our model runs at 56.8 FPS and 48.6 FPS on the latest Android and iPhone mobile devices, respectively. To the best of our knowledge, this is the first real-time deep LSD available on mobile devices.

One of the main purposes of deep metric learning is to construct an embedding space that has well-generalized embeddings on both seen (training) classes and unseen (test) classes. Most existing works have tried to achieve this using different types of metric objectives and hard sample mining strategies with given training data. However, learning with only the training data can be overfitted to the seen classes, leading to the lack of generalization capability on unseen classes. To address this problem, we propose a simple regularizer called Proxy Synthesis that exploits synthetic classes for stronger generalization in deep metric learning. The proposed method generates synthetic embeddings and proxies that work as synthetic classes, and they mimic unseen classes when computing proxy-based losses. Proxy Synthesis derives an embedding space considering class relations and smooth decision boundaries for robustness on unseen classes. Our method is applicable to any proxy-based losses, including softmax and its variants. Extensive experiments on four famous benchmarks in image retrieval tasks demonstrate that Proxy Synthesis significantly boosts the performance of proxy-based losses and achieves state-of-the-art performance. Our implementation is available at github.com/navervision/proxy-synthesis.

The core of deep metric learning (DML) involves learning visual similarities in high-dimensional embedding space. One of the main challenges is to generalize from seen classes of training data to unseen classes of test data. Recent works have focused on exploiting past embeddings to increase the number of instances for the seen classes. Such methods achieve performance improvement via augmentation, while the strong focus on seen classes still remains. This can be undesirable for DML, where training and test data exhibit entirely different classes. In this work, we present a novel training strategy for DML called MemVir. Unlike previous works, MemVir memorizes both embedding features and class weights to utilize them as additional virtual classes. The exploitation of virtual classes not only utilizes augmented information for training but also alleviates a strong focus on seen classes for better generalization. Moreover, we embed the idea of curriculum learning by slowly adding virtual classes for a gradual increase in learning difficulty, which improves the learning stability as well as the final performance. MemVir can be easily applied to many existing loss functions without any modification. Extensive experimental results on famous benchmarks demonstrate the superiority of MemVir over state-of-the-art competitors. Code of MemVir is publicly available1.

Deep metric learning aims to learn embeddings that contain semantic similarity information among data points. To learn better embeddings, methods to generate synthetic hard samples have been proposed. Existing methods of synthetic hard sample generation are adopting autoencoders or generative adversarial networks, but this leads to more hyper-parameters, harder optimization, and slower training speed. In this paper, we address these problems by proposing a novel method of synthetic hard sample generation called symmetrical synthesis. Given two original feature points from the same class, the proposed method firstly generates synthetic points with each other as an axis of symmetry. Secondly, it performs hard negative pair mining within the original and synthetic points to select a more informative negative pair for computing the metric learning loss. Our proposed method is hyper-parameter free and plug-and-play for existing metric learning losses without network modification. We demonstrate the superiority of our proposed method over existing methods for a variety of loss functions on clustering and image retrieval tasks.

Learning the distance metric between pairs of samples has been studied for image retrieval and clustering. With the remarkable success of pair-based metric learning losses, recent works have proposed the use of generated synthetic points on metric learning losses for augmentation and generalization. However, these methods require additional generative networks along with the main network, which can lead to a larger model size, slower training speed, and harder optimization. Meanwhile, post-processing techniques, such as query expansion and database augmentation, have proposed the combination of feature points to obtain additional semantic information. In this paper, inspired by query expansion and database augmentation, we propose an augmentation method in an embedding space for pair-based metric learning losses, called embedding expansion. The proposed method generates synthetic points containing augmented information by a combination of feature points and performs hard negative pair mining to learn with the most informative feature representations. Because of its simplicity and flexibility, it can be used for existing metric learning losses without affecting model size, training speed, or optimization difficulty. Finally, the combination of embedding expansion and representative metric learning losses outperforms the state-of-the-art losses and previous sample generation methods in both image retrieval and clustering tasks. The implementation is publicly available.

There are early studies to attempt users for psychiatric counseling with chatbot. They lead to changes in drinking habit based on intervention approach via chat bot. The application does not consider the user's psychiatric status through the conversations, continuous user monitoring, and ethical judgment in the intervention. We contend that more accurate and continuous emotion recognition gives better satisfaction to users who need mental health care. In addition, appropriate clinical psychological response based on ethical responses is as well. We suggest a conversational service for psychiatric counseling that is adapted methodologies to understand counseling contents based on of high-level natural language understanding (NLU), and emotion recognition based on multi-modal approach. The methodologies enable continuous observation of emotional changes sensitively. In addition, the case-based counseling response model that combines ethical judgment model provides a suitable response to clinical psychiatric counseling.

Dropout is a technique widely used for preventing overfitting while training deep neural networks. However, applying dropout to a neural network typically increases the training time. This paper proposes a different dropout approach called controlled dropout that improves training speed by dropping units in a column-wise or row-wise manner on the matrices. In controlled dropout, a network is trained using compressed matrices of smaller size, which results in notable improvement of training speed. In the experiment on feed-forward neural networks for MNIST data set and convolutional neural networks for CIFAR-10 and SVHN data sets, our proposed method achieves faster training speed than conventional methods both on CPU and GPU, while exhibiting the same regularization performance as conventional dropout. Moreover, the improvement of training speed increases when the number of fully-connected layers increases. As the training process of neural network is an iterative process comprising forward propagation and backpropagation, speed improvement using controlled dropout would provide a significantly decreased training time.