Efficient Adaptation of Pre-trained Vision Transformer via Householder Transformation

A common strategy for Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViTs) involves adapting the model to downstream tasks by learning a low-rank adaptation matrix. This matrix is decomposed into a product of down-projection and up-projection matrices, with the bottleneck dimensionality being crucial for reducing the number of learnable parameters, as exemplified by prevalent methods like LoRA and Adapter. However, these low-rank strategies typically employ a fixed bottleneck dimensionality, which limits their flexibility in handling layer-wise variations. To address this limitation, we propose a novel PEFT approach inspired by Singular Value Decomposition (SVD) for representing the adaptation matrix. SVD decomposes a matrix into the product of a left unitary matrix, a diagonal matrix of scaling values, and a right unitary matrix. We utilize Householder transformations to construct orthogonal matrices that efficiently mimic the unitary matrices, requiring only a vector. The diagonal values are learned in a layer-wise manner, allowing them to flexibly capture the unique properties of each layer. This approach enables the generation of adaptation matrices with varying ranks across different layers, providing greater flexibility in adapting pre-trained models. Experiments on standard downstream vision tasks demonstrate that our method achieves promising fine-tuning performance.

AutoTVG: A New Vision-language Pre-training Paradigm for Temporal Video Grounding

Temporal Video Grounding (TVG) aims to localize a moment from an untrimmed video given the language description. Since the annotation of TVG is labor-intensive, TVG under limited supervision has accepted attention in recent years. The great success of vision-language pre-training guides TVG to follow the traditional "pre-training + fine-tuning" paradigm, however, the pre-training process would suffer from a lack of temporal modeling and fine-grained alignment due to the difference of data nature between pre-train and test. Besides, the large gap between pretext and downstream tasks makes zero-shot testing impossible for the pre-trained model. To avoid the drawbacks of the traditional paradigm, we propose AutoTVG, a new vision-language pre-training paradigm for TVG that enables the model to learn semantic alignment and boundary regression from automatically annotated untrimmed videos. To be specific, AutoTVG consists of a novel Captioned Moment Generation (CMG) module to generate captioned moments from untrimmed videos, and TVGNet with a regression head to predict localization results. Experimental results on Charades-STA and ActivityNet Captions show that, regarding zero-shot temporal video grounding, AutoTVG achieves highly competitive performance with in-distribution methods under out-of-distribution testing, and is superior to existing pre-training frameworks with much less training data.

Logic Synthesis with Generative Deep Neural Networks

While deep learning has achieved significant success in various domains, its application to logic circuit design has been limited due to complex constraints and strict feasibility requirement. However, a recent generative deep neural model, "Circuit Transformer", has shown promise in this area by enabling equivalence-preserving circuit transformation on a small scale. In this paper, we introduce a logic synthesis rewriting operator based on the Circuit Transformer model, named "ctrw" (Circuit Transformer Rewriting), which incorporates the following techniques: (1) a two-stage training scheme for the Circuit Transformer tailored for logic synthesis, with iterative improvement of optimality through self-improvement training; (2) integration of the Circuit Transformer with state-of-the-art rewriting techniques to address scalability issues, allowing for guided DAG-aware rewriting. Experimental results on the IWLS 2023 contest benchmark demonstrate the effectiveness of our proposed rewriting methods.

Low-Rank Rescaled Vision Transformer Fine-Tuning: A Residual Design Approach

Parameter-efficient fine-tuning for pre-trained Vision Transformers aims to adeptly tailor a model to downstream tasks by learning a minimal set of new adaptation parameters while preserving the frozen majority of pre-trained parameters. Striking a balance between retaining the generalizable representation capacity of the pre-trained model and acquiring task-specific features poses a key challenge. Currently, there is a lack of focus on guiding this delicate trade-off. In this study, we approach the problem from the perspective of Singular Value Decomposition (SVD) of pre-trained parameter matrices, providing insights into the tuning dynamics of existing methods. Building upon this understanding, we propose a Residual-based Low-Rank Rescaling (RLRR) fine-tuning strategy. This strategy not only enhances flexibility in parameter tuning but also ensures that new parameters do not deviate excessively from the pre-trained model through a residual design. Extensive experiments demonstrate that our method achieves competitive performance across various downstream image classification tasks, all while maintaining comparable new parameters. We believe this work takes a step forward in offering a unified perspective for interpreting existing methods and serves as motivation for the development of new approaches that move closer to effectively considering the crucial trade-off mentioned above. Our code is available at \href{https://github.com/zstarN70/RLRR.git}{https://github.com/zstarN70/RLRR.git}.

Circuit Transformer: End-to-end Circuit Design by Predicting the Next Gate

Language, a prominent human ability to express through sequential symbols, has been computationally mastered by recent advances of large language models (LLMs). By predicting the next word recurrently with huge neural models, LLMs have shown unprecedented capabilities in understanding and reasoning. Circuit, as the "language" of electronic design, specifies the functionality of an electronic device by cascade connections of logic gates. Then, can circuits also be mastered by a a sufficiently large "circuit model", which can conquer electronic design tasks by simply predicting the next logic gate? In this work, we take the first step to explore such possibilities. Two primary barriers impede the straightforward application of LLMs to circuits: their complex, non-sequential structure, and the intolerance of hallucination due to strict constraints (e.g., equivalence). For the first barrier, we encode a circuit as a memory-less, depth-first traversal trajectory, which allows Transformer-based neural models to better leverage its structural information, and predict the next gate on the trajectory as a circuit model. For the second barrier, we introduce an equivalence-preserving decoding process, which ensures that every token in the generated trajectory adheres to the specified equivalence constraints. Moreover, the circuit model can also be regarded as a stochastic policy to tackle optimization-oriented circuit design tasks. Experimentally, we trained a Transformer-based model of 88M parameters, named "Circuit Transformer", which demonstrates impressive performance in end-to-end logic synthesis. With Monte-Carlo tree search, Circuit Transformer significantly improves over resyn2 while retaining strict equivalence, showcasing the potential of generative AI in conquering electronic design challenges.

Joint Beam Scheduling and Power Optimization for Beam Hopping LEO Satellite Systems

Low earth orbit (LEO) satellite communications can provide ubiquitous and reliable services, making it an essential part of the Internet of Everything network. Beam hopping (BH) is an emerging technology for effectively addressing the issue of low resource utilization caused by the non-uniform spatio-temporal distribution of traffic demands. However, how to allocate multi-dimensional resources in a timely and efficient way for the highly dynamic LEO satellite systems remains a challenge. This paper proposes a joint beam scheduling and power optimization beam hopping (JBSPO-BH) algorithm considering the differences in the geographic distribution of sink nodes. The JBSPO-BH algorithm decouples the original problem into two sub-problems. The beam scheduling problem is modelled as a potential game, and the Nash equilibrium (NE) point is obtained as the beam scheduling strategy. Moreover, the penalty function interior point method is applied to optimize the power allocation. Simulation results show that the JBSPO-BH algorithm has low time complexity and fast convergence and achieves better performance both in throughput and fairness. Compared with greedy-based BH, greedy-based BH with the power optimization, round-robin BH, Max-SINR BH and satellite resource allocation algorithm, the throughput of the proposed algorithm is improved by 44.99%, 20.79%, 156.06%, 15.39% and 8.17%, respectively.

VideoAssembler: Identity-Consistent Video Generation with Reference Entities using Diffusion Model

Identity-consistent video generation seeks to synthesize videos that are guided by both textual prompts and reference images of entities. Current approaches typically utilize cross-attention layers to integrate the appearance of the entity, which predominantly captures semantic attributes, resulting in compromised fidelity of entities. Moreover, these methods necessitate iterative fine-tuning for each new entity encountered, thereby limiting their applicability. To address these challenges, we introduce VideoAssembler, a novel end-to-end framework for identity-consistent video generation that can conduct inference directly when encountering new entities. VideoAssembler is adept at producing videos that are not only flexible with respect to the input reference entities but also responsive to textual conditions. Additionally, by modulating the quantity of input images for the entity, VideoAssembler enables the execution of tasks ranging from image-to-video generation to sophisticated video editing. VideoAssembler comprises two principal components: the Reference Entity Pyramid (REP) encoder and the Entity-Prompt Attention Fusion (EPAF) module. The REP encoder is designed to infuse comprehensive appearance details into the denoising stages of the stable diffusion model. Concurrently, the EPAF module is utilized to integrate text-aligned features effectively. Furthermore, to mitigate the challenge of scarce data, we present a methodology for the preprocessing of training data. Our evaluation of the VideoAssembler framework on the UCF-101, MSR-VTT, and DAVIS datasets indicates that it achieves good performances in both quantitative and qualitative analyses (346.84 in FVD and 48.01 in IS on UCF-101). Our project page is at https://gulucaptain.github.io/videoassembler/.

Communication Efficiency Optimization of Federated Learning for Computing and Network Convergence of 6G Networks

Federated learning effectively addresses issues such as data privacy by collaborating across participating devices to train global models. However, factors such as network topology and device computing power can affect its training or communication process in complex network environments. A new network architecture and paradigm with computing-measurable, perceptible, distributable, dispatchable, and manageable capabilities, computing and network convergence (CNC) of 6G networks can effectively support federated learning training and improve its communication efficiency. By guiding the participating devices' training in federated learning based on business requirements, resource load, network conditions, and arithmetic power of devices, CNC can reach this goal. In this paper, to improve the communication efficiency of federated learning in complex networks, we study the communication efficiency optimization of federated learning for computing and network convergence of 6G networks, methods that gives decisions on its training process for different network conditions and arithmetic power of participating devices in federated learning. The experiments address two architectures that exist for devices in federated learning and arrange devices to participate in training based on arithmetic power while achieving optimization of communication efficiency in the process of transferring model parameters. The results show that the method we proposed can (1) cope well with complex network situations (2) effectively balance the delay distribution of participating devices for local training (3) improve the communication efficiency during the transfer of model parameters (4) improve the resource utilization in the network.

Fuse Your Latents: Video Editing with Multi-source Latent Diffusion Models

Latent Diffusion Models (LDMs) are renowned for their powerful capabilities in image and video synthesis. Yet, video editing methods suffer from insufficient pre-training data or video-by-video re-training cost. In addressing this gap, we propose FLDM (Fused Latent Diffusion Model), a training-free framework to achieve text-guided video editing by applying off-the-shelf image editing methods in video LDMs. Specifically, FLDM fuses latents from an image LDM and an video LDM during the denoising process. In this way, temporal consistency can be kept with video LDM while high-fidelity from the image LDM can also be exploited. Meanwhile, FLDM possesses high flexibility since both image LDM and video LDM can be replaced so advanced image editing methods such as InstructPix2Pix and ControlNet can be exploited. To the best of our knowledge, FLDM is the first method to adapt off-the-shelf image editing methods into video LDMs for video editing. Extensive quantitative and qualitative experiments demonstrate that FLDM can improve the textual alignment and temporal consistency of edited videos.

Reuse and Diffuse: Iterative Denoising for Text-to-Video Generation

Inspired by the remarkable success of Latent Diffusion Models (LDMs) for image synthesis, we study LDM for text-to-video generation, which is a formidable challenge due to the computational and memory constraints during both model training and inference. A single LDM is usually only capable of generating a very limited number of video frames. Some existing works focus on separate prediction models for generating more video frames, which suffer from additional training cost and frame-level jittering, however. In this paper, we propose a framework called "Reuse and Diffuse" dubbed $\textit{VidRD}$ to produce more frames following the frames already generated by an LDM. Conditioned on an initial video clip with a small number of frames, additional frames are iteratively generated by reusing the original latent features and following the previous diffusion process. Besides, for the autoencoder used for translation between pixel space and latent space, we inject temporal layers into its decoder and fine-tune these layers for higher temporal consistency. We also propose a set of strategies for composing video-text data that involve diverse content from multiple existing datasets including video datasets for action recognition and image-text datasets. Extensive experiments show that our method achieves good results in both quantitative and qualitative evaluations. Our project page is available $\href{https://anonymous0x233.github.io/ReuseAndDiffuse/}{here}$.