Graph Transformers, emerging as a new architecture for graph representation learning, suffer from the quadratic complexity on the number of nodes when handling large graphs. To this end, we propose a Neighborhood Aggregation Graph Transformer (NAGphormer) that treats each node as a sequence containing a series of tokens constructed by our proposed Hop2Token module. For each node, Hop2Token aggregates the neighborhood features from different hops into different representations, producing a sequence of token vectors as one input. In this way, NAGphormer could be trained in a mini-batch manner and thus could scale to large graphs. Moreover, we mathematically show that compared to a category of advanced Graph Neural Networks (GNNs), called decoupled Graph Convolutional Networks, NAGphormer could learn more informative node representations from multi-hop neighborhoods. In addition, we propose a new data augmentation method called Neighborhood Augmentation (NrAug) based on the output of Hop2Token that augments simultaneously the features of neighborhoods from global as well as local views to strengthen the training effect of NAGphormer. Extensive experiments on benchmark datasets from small to large demonstrate the superiority of NAGphormer against existing graph Transformers and mainstream GNNs, and the effectiveness of NrAug for further boosting NAGphormer.
Text-video retrieval is a challenging cross-modal task, which aims to align visual entities with natural language descriptions. Current methods either fail to leverage the local details or are computationally expensive. What's worse, they fail to leverage the heterogeneous concepts in data. In this paper, we propose the Disentangled Conceptualization and Set-to-set Alignment (DiCoSA) to simulate the conceptualizing and reasoning process of human beings. For disentangled conceptualization, we divide the coarse feature into multiple latent factors related to semantic concepts. For set-to-set alignment, where a set of visual concepts correspond to a set of textual concepts, we propose an adaptive pooling method to aggregate semantic concepts to address the partial matching. In particular, since we encode concepts independently in only a few dimensions, DiCoSA is superior at efficiency and granularity, ensuring fine-grained interactions using a similar computational complexity as coarse-grained alignment. Extensive experiments on five datasets, including MSR-VTT, LSMDC, MSVD, ActivityNet, and DiDeMo, demonstrate that our method outperforms the existing state-of-the-art methods.
Diffusion models, either with or without text condition, have demonstrated impressive capability in synthesizing photorealistic images given a few or even no words. These models may not fully satisfy user need, as normal users or artists intend to control the synthesized images with specific guidance, like overall layout, color, structure, object shape, and so on. To adapt diffusion models for controllable image synthesis, several methods have been proposed to incorporate the required conditions as regularization upon the intermediate features of the diffusion denoising network. These methods, known as early-constraint ones in this paper, have difficulties in handling multiple conditions with a single solution. They intend to train separate models for each specific condition, which require much training cost and result in non-generalizable solutions. To address these difficulties, we propose a new approach namely late-constraint: we leave the diffusion networks unchanged, but constrain its output to be aligned with the required conditions. Specifically, we train a lightweight condition adapter to establish the correlation between external conditions and internal representations of diffusion models. During the iterative denoising process, the conditional guidance is sent into corresponding condition adapter to manipulate the sampling process with the established correlation. We further equip the introduced late-constraint strategy with a timestep resampling method and an early stopping technique, which boost the quality of synthesized image meanwhile complying with the guidance. Our method outperforms the existing early-constraint methods and generalizes better to unseen condition.
Video question answering aims at answering a question about the video content by reasoning the alignment semantics within them. However, since relying heavily on human instructions, i.e., annotations or priors, current contrastive learning-based VideoQA methods remains challenging to perform fine-grained visual-linguistic alignments. In this work, we innovatively resort to game theory, which can simulate complicated relationships among multiple players with specific interaction strategies, e.g., video, question, and answer as ternary players, to achieve fine-grained alignment for VideoQA task. Specifically, we carefully design a VideoQA-specific interaction strategy to tailor the characteristics of VideoQA, which can mathematically generate the fine-grained visual-linguistic alignment label without label-intensive efforts. Our TG-VQA outperforms existing state-of-the-art by a large margin (more than 5%) on long-term and short-term VideoQA datasets, verifying its effectiveness and generalization ability. Thanks to the guidance of game-theoretic interaction, our model impressively convergences well on limited data (${10}^4 ~videos$), surpassing most of those pre-trained on large-scale data ($10^7~videos$).
Weakly supervised learning based on scribble annotations in target extraction of remote sensing images has drawn much interest due to scribbles' flexibility in denoting winding objects and low cost of manually labeling. However, scribbles are too sparse to identify object structure and detailed information, bringing great challenges in target localization and boundary description. To alleviate these problems, in this paper, we construct two inner structure-constraints, a deformation consistency loss and a trainable active contour loss, together with a scribble-constraint to supervise the optimization of the encoder-decoder network without introducing any auxiliary module or extra operation based on prior cues. Comprehensive experiments demonstrate our method's superiority over five state-of-the-art algorithms in this field. Source code is available at https://github.com/yitongli123/ISC-TE.
LLMs now exhibit human-like skills in various fields, leading to worries about misuse. Thus, detecting generated text is crucial. However, passive detection methods are stuck in domain specificity and limited adversarial robustness. To achieve reliable detection, a watermark-based method was proposed for white-box LLMs, allowing them to embed watermarks during text generation. The method involves randomly dividing the model vocabulary to obtain a special list and adjusting the probability distribution to promote the selection of words in the list. A detection algorithm aware of the list can identify the watermarked text. However, this method is not applicable in many real-world scenarios where only black-box language models are available. For instance, third-parties that develop API-based vertical applications cannot watermark text themselves because API providers only supply generated text and withhold probability distributions to shield their commercial interests. To allow third-parties to autonomously inject watermarks into generated text, we develop a watermarking framework for black-box language model usage scenarios. Specifically, we first define a binary encoding function to compute a random binary encoding corresponding to a word. The encodings computed for non-watermarked text conform to a Bernoulli distribution, wherein the probability of a word representing bit-1 being approximately 0.5. To inject a watermark, we alter the distribution by selectively replacing words representing bit-0 with context-based synonyms that represent bit-1. A statistical test is then used to identify the watermark. Experiments demonstrate the effectiveness of our method on both Chinese and English datasets. Furthermore, results under re-translation, polishing, word deletion, and synonym substitution attacks reveal that it is arduous to remove the watermark without compromising the original semantics.
Stable or metastable crystal structures of assembled atoms can be predicted by finding the global or local minima of the energy surface with respect to the atomic configurations. Generally, this requires repeated first-principles energy calculations that are impractical for large systems, such as those containing more than 30 atoms in the unit cell. Here, we have made significant progress in solving the crystal structure prediction problem with a simple but powerful machine-learning workflow; using a machine-learning surrogate for first-principles energy calculations, we performed non-iterative, single-shot screening using a large library of virtually created crystal structures. The present method relies on two key technical components: transfer learning, which enables a highly accurate energy prediction of pre-relaxed crystalline states given only a small set of training samples from first-principles calculations, and generative models to create promising and diverse crystal structures for screening. Here, first-principles calculations were performed only to generate the training samples, and for the optimization of a dozen or fewer finally narrowed-down crystal structures. Our shotgun method was more than 5--10 times less computationally demanding and achieved an outstanding prediction accuracy that was 2--6 times higher than that of the conventional methods that rely heavily on iterative first-principles calculations.
Integrated sensing and communication (ISAC) has recently merged as a promising technique to provide sensing services in future wireless networks. In the literature, numerous works have adopted a monostatic radar architecture to realize ISAC, i.e., employing the same base station (BS) to transmit the ISAC signal and receive the echo. Yet, the concurrent information transmission causes severe self-interference (SI) to the radar echo at the BS which cannot be effectively suppressed. To overcome this difficulty, in this paper, we propose a coordinated cellular network-supported multistatic radar architecture to implement ISAC. In particular, among all the coordinated BSs, we select a BS as the multistatic receiver to receive the sensing echo signal, while the other BSs act as the multistatic transmitters to collaborate with each other to facilitate cooperative ISAC. This allows us to spatially separate the ISAC signal transmission and radar echo reception, intrinsically circumventing the problem of SI. To this end, we jointly optimize the transmit and receive beamforming policy to minimize the sensing beam pattern mismatch error subject to both the communication and sensing quality-of-service requirements. The resulting non-convex optimization problem is tackled by a low-complexity alternating optimization-based suboptimal algorithm. Simulation results showed that the proposed scheme outperforms the two baseline schemes adopting conventional designs. Moreover, our results confirm that the proposed architecture is promising in achieving high-quality ISAC.
To guarantee excellent reliability performance in ultra-reliable low-latency communications (URLLC), pragmatic precoder design is an effective approach. However, an efficient precoder design highly depends on the accurate instantaneous channel state information at the transmitter (ICSIT), which however, is not always available in practice. To overcome this problem, in this paper, we focus on the orthogonal time frequency space (OTFS)-based URLLC system and adopt a deep learning (DL) approach to directly predict the precoder for the next time frame to minimize the frame error rate (FER) via implicitly exploiting the features from estimated historical channels in the delay-Doppler domain. By doing this, we can guarantee the system reliability even without the knowledge of ICSIT. To this end, a general precoder design problem is formulated where a closed-form theoretical FER expression is specifically derived to characterize the system reliability. Then, a delay-Doppler domain channels-aware convolutional long short-term memory (CLSTM) network (DDCL-Net) is proposed for predictive precoder design. In particular, both the convolutional neural network and LSTM modules are adopted in the proposed neural network to exploit the spatial-temporal features of wireless channels for improving the learning performance. Finally, simulation results demonstrated that the FER performance of the proposed method approaches that of the perfect ICSI-aided scheme.