Existing semi-supervised learning (SSL) methods assume that labeled and unlabeled data share the same class space. However, in real-world applications, unlabeled data always contain classes not present in the labeled set, which may cause classification performance degradation of known classes. Therefore, open-world SSL approaches are researched to handle the presence of multiple unknown classes in the unlabeled data, which aims to accurately classify known classes while fine-grained distinguishing different unknown classes. To address this challenge, in this paper, we propose an open-world SSL method for Self-learning Open-world Classes (SSOC), which can explicitly self-learn multiple unknown classes. Specifically, SSOC first defines class center tokens for both known and unknown classes and autonomously learns token representations according to all samples with the cross-attention mechanism. To effectively discover novel classes, SSOC further designs a pairwise similarity loss in addition to the entropy loss, which can wisely exploit the information available in unlabeled data from instances' predictions and relationships. Extensive experiments demonstrate that SSOC outperforms the state-of-the-art baselines on multiple popular classification benchmarks. Specifically, on the ImageNet-100 dataset with a novel ratio of 90%, SSOC achieves a remarkable 22% improvement.
In this paper, we introduce "InfiAgent-DABench", the first benchmark specifically designed to evaluate LLM-based agents in data analysis tasks. This benchmark contains DAEval, a dataset consisting of 311 data analysis questions derived from 55 CSV files, and an agent framework to evaluate LLMs as data analysis agents. We adopt a format-prompting technique, ensuring questions to be closed-form that can be automatically evaluated. Our extensive benchmarking of 23 state-of-the-art LLMs uncovers the current challenges encountered in data analysis tasks. In addition, we have developed DAAgent, a specialized agent trained on instruction-tuning datasets. Evaluation datasets and toolkits for InfiAgent-DABench are released at https://github.com/InfiAgent/InfiAgent.
Striking a balance between precision and efficiency presents a prominent challenge in the bird's-eye-view (BEV) 3D object detection. Although previous camera-based BEV methods achieved remarkable performance by incorporating long-term temporal information, most of them still face the problem of low efficiency. One potential solution is knowledge distillation. Existing distillation methods only focus on reconstructing spatial features, while overlooking temporal knowledge. To this end, we propose TempDistiller, a Temporal knowledge Distiller, to acquire long-term memory from a teacher detector when provided with a limited number of frames. Specifically, a reconstruction target is formulated by integrating long-term temporal knowledge through self-attention operation applied to feature teachers. Subsequently, novel features are generated for masked student features via a generator. Ultimately, we utilize this reconstruction target to reconstruct the student features. In addition, we also explore temporal relational knowledge when inputting full frames for the student model. We verify the effectiveness of the proposed method on the nuScenes benchmark. The experimental results show our method obtain an enhancement of +1.6 mAP and +1.1 NDS compared to the baseline, a speed improvement of approximately 6 FPS after compressing temporal knowledge, and the most accurate velocity estimation.
In this paper, we propose a novel cascaded diffusion-based generative framework for text-driven human motion synthesis, which exploits a strategy named GradUally Enriching SyntheSis (GUESS as its abbreviation). The strategy sets up generation objectives by grouping body joints of detailed skeletons in close semantic proximity together and then replacing each of such joint group with a single body-part node. Such an operation recursively abstracts a human pose to coarser and coarser skeletons at multiple granularity levels. With gradually increasing the abstraction level, human motion becomes more and more concise and stable, significantly benefiting the cross-modal motion synthesis task. The whole text-driven human motion synthesis problem is then divided into multiple abstraction levels and solved with a multi-stage generation framework with a cascaded latent diffusion model: an initial generator first generates the coarsest human motion guess from a given text description; then, a series of successive generators gradually enrich the motion details based on the textual description and the previous synthesized results. Notably, we further integrate GUESS with the proposed dynamic multi-condition fusion mechanism to dynamically balance the cooperative effects of the given textual condition and synthesized coarse motion prompt in different generation stages. Extensive experiments on large-scale datasets verify that GUESS outperforms existing state-of-the-art methods by large margins in terms of accuracy, realisticness, and diversity. Code is available at https://github.com/Xuehao-Gao/GUESS.
We present StreamVC, a streaming voice conversion solution that preserves the content and prosody of any source speech while matching the voice timbre from any target speech. Unlike previous approaches, StreamVC produces the resulting waveform at low latency from the input signal even on a mobile platform, making it applicable to real-time communication scenarios like calls and video conferencing, and addressing use cases such as voice anonymization in these scenarios. Our design leverages the architecture and training strategy of the SoundStream neural audio codec for lightweight high-quality speech synthesis. We demonstrate the feasibility of learning soft speech units causally, as well as the effectiveness of supplying whitened fundamental frequency information to improve pitch stability without leaking the source timbre information.
Semantic communication (SemCom) has received considerable attention for its ability to reduce data transmission size while maintaining task performance. However, existing works mainly focus on analog SemCom with simple channel models, which may limit its practical application. To reduce this gap, we propose an orthogonal frequency division multiplexing (OFDM)-based SemCom system that is compatible with existing digital communication infrastructures. In the considered system, the extracted semantics is quantized by scalar quantizers, transformed into OFDM signal, and then transmitted over the frequency-selective channel. Moreover, we propose a semantic importance measurement method to build the relationship between target task and semantic features. Based on semantic importance, we formulate a sub-carrier and bit allocation problem to maximize communication performance. However, the optimization objective function cannot be accurately characterized using a mathematical expression due to the neural network-based semantic codec. Given the complex nature of the problem, we first propose a low-complexity sub-carrier allocation method that assigns sub-carriers with better channel conditions to more critical semantics. Then, we propose a deep reinforcement learning-based bit allocation algorithm with dynamic action space. Simulation results demonstrate that the proposed system achieves 9.7% and 28.7% performance gains compared to analog SemCom and conventional bit-based communication systems, respectively.
Flagellated microorganisms can swim at low Reynolds numbers and adapt to changes in their environment. Specifically, the flagella can switch their shapes or modes through gene expression. In the past decade, efforts have been made to fabricate and investigate rigid types of microrobots without any adaptation to the environments. More recently, obtaining adaptive microrobots mimicking real microorganisms is getting more attention. However, even though some adaptive microrobots achieved by hydrogels have emerged, the swimming behaviors of the microrobots before and after the environment-induced deformations are not predicted in a systematic standardized way. In this work, experiments, finite element analysis, and dynamic modeling are presented together to realize a complete understanding of these adaptive microrobots. The above three parts are cross-verified proving the success of using such methods, facilitating the bio-applications with shape-programmable and even swimming performance-programmable microrobots. Moreover, an application of targeted object delivery using the proposed microrobot has been successfully demonstrated. Finally, cytotoxicity tests are performed to prove the potential for using the proposed microrobot for biomedical applications.
We present MobileVLM, a competent multimodal vision language model (MMVLM) targeted to run on mobile devices. It is an amalgamation of a myriad of architectural designs and techniques that are mobile-oriented, which comprises a set of language models at the scale of 1.4B and 2.7B parameters, trained from scratch, a multimodal vision model that is pre-trained in the CLIP fashion, cross-modality interaction via an efficient projector. We evaluate MobileVLM on several typical VLM benchmarks. Our models demonstrate on par performance compared with a few much larger models. More importantly, we measure the inference speed on both a Qualcomm Snapdragon 888 CPU and an NVIDIA Jeston Orin GPU, and we obtain state-of-the-art performance of 21.5 tokens and 65.3 tokens per second, respectively. Our code will be made available at: https://github.com/Meituan-AutoML/MobileVLM.