arXiv subjects

Shichao Li

Publications and source records attributed to Shichao Li.

At least 19 recordsLinked to original sources

arXiv2026

Compiling and Benchmarking Task-State Horizons for Embodied Agents

Frontier agentic models are increasingly deployed as high-level planners for long-horizon embodied tasks. Existing robotic benchmarks have advanced long-horizon evaluation, but primarily characterize difficulty through action-sequence length and subtask complexity, overlooking a distinct challenge: agents must track evolving task-relevant world states induced by both their exploration and environmental dynamics. We define the span of task-relevant state transitions that an agent must track as task-state horizon (TSH). To evaluate how agent performance varies with TSH, we introduce RoboGraph, a robotic task compiler that translates state-transition dependencies into executable symbolic graphs. Specifically, RoboGraph constructs task-state horizons from spatial and temporal causal dependencies, including those induced by unexpected failures and interventions during task execution. Building on RoboGraph, we release a benchmark comprising 588 episodes across 84 scenes with varying TSHs. Experiments evaluating 15 advanced agentic models in both semantic and visual closed-loop environments show that most models struggle with demanding TSHs, revealing substantial gaps in maintaining, exploring, and updating task-relevant state over long horizon.

cs.RO↗

arXiv2026

Toward Deep Representation Learning for Event-Enhanced Visual Autonomous Perception: the eAP Dataset

Recent visual autonomous perception systems achieve remarkable performances with deep representation learning. However, they fail in scenarios with challenging illumination.While event cameras can mitigate this problem, there is a lack of a large-scale dataset to develop event-enhanced deep visual perception models in autonomous driving scenes. To address the gap, we present the eAP (event-enhanced Autonomous Perception) dataset, the largest dataset with event cameras for autonomous perception. We demonstrate how eAP can facilitate the study of different autonomous perception tasks, including 3D vehicle detection and object time-to-contact (TTC) estimation, through deep representation learning. Based on eAP, we demonstrate the ffrst successful use of events to improve a popular 3D vehicle detection network in challenging illumination scenarios. eAP also enables a devoted study of the representation learning problem of object TTC estimation. We show how a geometryaware representation learning framework leads to the best eventbased object TTC estimation network that operates at 200 FPS. The dataset, code, and pre-trained models will be made publicly available for future research.

cs.RO↗

arXiv2026

Lightweight Edge Learning via Dataset Pruning

Edge learning facilitates ubiquitous intelligence by enabling model training and adaptation directly on data-generating devices, thereby mitigating privacy risks and communication latency. However, the high computational and energy overhead of on-device training hinders its deployment on battery-powered mobile systems with strict thermal and memory budgets. While prior research has extensively optimized model architectures for efficient inference, the training phase remains bottlenecked by the processing of massive, often redundant, local datasets. In this work, we propose a data-centric optimization framework that leverages dataset pruning to achieve resource-efficient edge learning. Unlike standard methods that process all available data, our approach constructs compact, highly informative training subsets via a lightweight, on-device importance evaluation. Specifically, we utilize average loss statistics derived from a truncated warm-up phase to rank sample importance, deterministically retaining only the most critical data points under a dynamic pruning ratio. This mechanism is model-agnostic and operates locally without inter-device communication. Extensive experiments on standard image classification benchmarks demonstrate that our framework achieves a near-linear reduction in training latency and energy consumption proportional to the pruning ratio, with negligible degradation in model accuracy. These results validate dataset pruning as a vital, complementary paradigm for enhancing the sustainability and scalability of learning on resource-constrained mobile edge devices.

cs.LG↗

arXiv2025

Learning better representations for crowded pedestrians in offboard LiDAR-camera 3D tracking-by-detection

Perceiving pedestrians in highly crowded urban environments is a difficult long-tail problem for learning-based autonomous perception. Speeding up 3D ground truth generation for such challenging scenes is performance-critical yet very challenging. The difficulties include the sparsity of the captured pedestrian point cloud and a lack of suitable benchmarks for a specific system design study. To tackle the challenges, we first collect a new multi-view LiDAR-camera 3D multiple-object-tracking benchmark of highly crowded pedestrians for in-depth analysis. We then build an offboard auto-labeling system that reconstructs pedestrian trajectories from LiDAR point cloud and multi-view images. To improve the generalization power for crowded scenes and the performance for small objects, we propose to learn high-resolution representations that are density-aware and relationship-aware. Extensive experiments validate that our approach significantly improves the 3D pedestrian tracking performance towards higher auto-labeling efficiency. The code will be publicly available at this HTTP URL.

cs.CV↗

arXiv2025

A Multi-Modal Deep Learning Framework for Pan-Cancer Prognosis

Prognostic task is of great importance as it closely related to the survival analysis of patients, the optimization of treatment plans and the allocation of resources. The existing prognostic models have shown promising results on specific datasets, but there are limitations in two aspects. On the one hand, they merely explore certain types of modal data, such as patient histopathology WSI and gene expression analysis. On the other hand, they adopt the per-cancer-per-model paradigm, which means the trained models can only predict the prognostic effect of a single type of cancer, resulting in weak generalization ability. In this paper, a deep-learning based model, named UMPSNet, is proposed. Specifically, to comprehensively understand the condition of patients, in addition to constructing encoders for histopathology images and genomic expression profiles respectively, UMPSNet further integrates four types of important meta data (demographic information, cancer type information, treatment protocols, and diagnosis results) into text templates, and then introduces a text encoder to extract textual features. In addition, the optimal transport OT-based attention mechanism is utilized to align and fuse features of different modalities. Furthermore, a guided soft mixture of experts (GMoE) mechanism is introduced to effectively address the issue of distribution differences among multiple cancer datasets. By incorporating the multi-modality of patient data and joint training, UMPSNet outperforms all SOTA approaches, and moreover, it demonstrates the effectiveness and generalization ability of the proposed learning paradigm of a single model for multiple cancer types. The code of UMPSNet is available at https://github.com/binging512/UMPSNet.

eess.IV↗

arXiv2024

Observation of quantum oscillations near the Mott-Ioffe-Regel limit in CaAs3

The Mott-Ioffe-Regel limit sets the lower bound of carrier mean free path for coherent quasiparticle transport. Metallicity beyond this limit is of great interest because it is often closely related to quantum criticality and unconventional superconductivity. Progress along this direction mainly focuses on the strange-metal behaviors originating from the evolution of quasiparticle scattering rate such as linear-in-temperature resistivity, while the quasiparticle coherence phenomena in this regime are much less explored due to the short mean free path at the diffusive bound. Here we report the observation of quantum oscillations from Landau quantization near the Mott-Ioffe-Regel limit in CaAs3. Despite the insulator-like temperature dependence of resistivity, CaAs3 presents giant magnetoresistance and prominent Shubnikov-de Haas oscillations from Fermi surfaces, indicating highly coherent band transport. In contrast, the quantum oscillation is absent in the magnetic torque. The quasiparticle effective mass increases systematically with magnetic fields, manifesting a much larger value than the expectation given by magneto-infrared spectroscopy. It suggests a strong many-body renormalization effect near Fermi surface. We find that these unconventional behaviors may be explained by the interplay between the mobility edge and the van Hove singularity, which results in the formation of coherent cyclotron orbits emerging at the diffusive bound. Our results call for further study on the electron correlation effect of the van Hove singularity.

cond-mat.str-el↗

arXiv2023

OpenDriver: An Open-Road Driver State Detection Dataset

Among numerous studies for driver state detection, wearable physiological measurements offer a practical method for real-time monitoring. However, there are few driver physiological datasets in open-road scenarios, and the existing datasets suffer from issues such as poor signal quality, small sample sizes, and short data collection periods. Therefore, in this paper, a large-scale multimodal driving dataset, OpenDriver, for driver state detection is developed. The OpenDriver encompasses a total of 3,278 driving trips, with a signal collection duration spanning approximately 4,600 hours. Two modalities of driving signals are enrolled in OpenDriver: electrocardiogram (ECG) signals and six-axis motion data of the steering wheel from a motion measurement unit (IMU), which were recorded from 81 drivers and their vehicles. Furthermore, three challenging tasks are involved in our work, namely ECG signal quality assessment, individual biometric identification based on ECG signals, and physiological signal analysis in complex driving environments. To facilitate research in these tasks, corresponding benchmarks have also been introduced. First, a noisy augmentation strategy is applied to generate a larger-scale ECG signal dataset with realistic noise simulation for quality assessment. Second, an end-to-end contrastive learning framework is employed for individual biometric identification. Finally, a comprehensive analysis of drivers' HRV features under different driving conditions is conducted. Each benchmark provides evaluation metrics and reference results. The OpenDriver dataset will be publicly available at https://github.com/bdne/OpenDriver.

cs.AI↗

arXiv2022

Aircraft Ground Taxiing Deduction and Conflict Early Warning Method Based on Control Command Information

Aircraft taxiing conflict is a threat to the safety of airport operations, mainly due to the human error in control command infor-mation. In order to solve the problem, The aircraft taxiing deduction and conflict early warning method based on control order information is proposed. This method does not need additional equipment and operating costs, and is completely based on his-torical data and control command information. When the aircraft taxiing command is given, the future route information will be deduced, and the probability of conflict with other taxiing aircraft will be calculated to achieve conflict detection and early warning of different levels. The method is validated by the aircraft taxi data from real airports. The results show that the method can effectively predict the aircraft taxiing process, and can provide early warning of possible conflicts. Due to the advantages of low cost and high accuracy, this method has the potential to be applied to airport operation decision support system.

eess.SY↗

arXiv2022

SDQ: Stochastic Differentiable Quantization with Mixed Precision

In order to deploy deep models in a computationally efficient manner, model quantization approaches have been frequently used. In addition, as new hardware that supports mixed bitwidth arithmetic operations, recent research on mixed precision quantization (MPQ) begins to fully leverage the capacity of representation by searching optimized bitwidths for different layers and modules in a network. However, previous studies mainly search the MPQ strategy in a costly scheme using reinforcement learning, neural architecture search, etc., or simply utilize partial prior knowledge for bitwidth assignment, which might be biased and sub-optimal. In this work, we present a novel Stochastic Differentiable Quantization (SDQ) method that can automatically learn the MPQ strategy in a more flexible and globally-optimized space with smoother gradient approximation. Particularly, Differentiable Bitwidth Parameters (DBPs) are employed as the probability factors in stochastic quantization between adjacent bitwidth choices. After the optimal MPQ strategy is acquired, we further train our network with entropy-aware bin regularization and knowledge distillation. We extensively evaluate our method for several networks on different hardware (GPUs and FPGA) and datasets. SDQ outperforms all state-of-the-art mixed or single precision quantization with a lower bitwidth and is even better than the full-precision counterparts across various ResNet and MobileNet families, demonstrating the effectiveness and superiority of our method.

cs.LG↗

arXiv2022

Enhanced low-energy magnetic excitations evidencing the Cu-induced localization in an Fe-based superconductor Fe$_{0.98}$Te$_{0.5}$Se$_{0.5}$

We have performed inelastic neutron scattering measurements on optimally-doped Fe$_{0.98}$Te$_{0.5}$Se$_{0.5}$ and 10% Cu-doped Fe$_{0.88}$Cu$_{0.1}$Te$_{0.5}$Se$_{0.5}$ to investigate the substitution effects on the spin excitations in the whole energy range up to 300 meV. It is found that substitution of Cu for Fe enhances the low-energy spin excitations ($\le$ 100 meV), especially around the (0.5, 0.5) point, and leaves the high-energy magnetic excitations intact. In contrast to the expectation that Cu with spin 1/2 will dilute the magnetic moments contributed by Fe with a larger spin, we find that the 10% Cu doping enlarges the effective fluctuating moment from 2.85 to 3.13 $μ_{\rm B}$/Fe, although there is no long- or short-range magnetic order around (0.5, 0.5) and (0.5, 0). The presence of enhanced magnetic excitations in the 10% Cu doped sample which is in the insulating state indicates that the magnetic excitations must have some contributions from the local moments, reflecting the dual nature of the magnetism in iron-based superconductors. We attribute the substitution effects to the localization of the itinerant electrons induced by Cu dopants. These results also indicate that the Cu doping does not act as electron donor as in a rigid-band shift model, but more as scattering centers that localize the system.

cond-mat.supr-con↗

arXiv2022

Stereo Neural Vernier Caliper

We propose a new object-centric framework for learning-based stereo 3D object detection. Previous studies build scene-centric representations that do not consider the significant variation among outdoor instances and thus lack the flexibility and functionalities that an instance-level model can offer. We build such an instance-level model by formulating and tackling a local update problem, i.e., how to predict a refined update given an initial 3D cuboid guess. We demonstrate how solving this problem can complement scene-centric approaches in (i) building a coarse-to-fine multi-resolution system, (ii) performing model-agnostic object location refinement, and (iii) conducting stereo 3D tracking-by-detection. Extensive experiments demonstrate the effectiveness of our approach, which achieves state-of-the-art performance on the KITTI benchmark. Code and pre-trained models are available at https://github.com/Nicholasli1995/SNVC.

cs.CV↗

arXiv2021

Automated Vision-Based Wellness Analysis for Elderly Care Centers

The growth in the aging population requires caregivers to improve both efficiency and quality of healthcare. In this study, we develop an automatic, vision-based system for monitoring and analyzing the physical and mental well-being of senior citizens. Through collaboration with Haven of Hope Christian Service, we collect video recording data in the care center with surveillance cameras. We then process and extract personalized facial, activity, and interaction features from the video data using deep neural networks. This integrated health information systems can assist caregivers to gain better insights into the seniors they are taking care of. These insights, including wellness metrics and long-term health patterns of senior citizens, can help caregivers update their caregiving strategies. We report the findings of our analysis and evaluate the system quantitatively. We also summarize technical challenges and additional functionalities and technologies needed for offering a comprehensive system.

cs.MM↗

arXiv2021

Joint stereo 3D object detection and implicit surface reconstruction

We present a new learning-based framework S-3D-RCNN that can recover accurate object orientation in SO(3) and simultaneously predict implicit rigid shapes from stereo RGB images. For orientation estimation, in contrast to previous studies that map local appearance to observation angles, we propose a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs). This approach features a deep model that transforms perceived intensities from one or two views to object part coordinates to achieve direct egocentric object orientation estimation in the camera coordinate system. To further achieve finer description inside 3D bounding boxes, we investigate the implicit shape estimation problem from stereo images. We model visible object surfaces by designing a point-based representation, augmenting IGRs to explicitly address the unseen surface hallucination problem. Extensive experiments validate the effectiveness of the proposed IGRs, and S-3D-RCNN achieves superior 3D scene understanding performance. We also designed new metrics on the KITTI benchmark for our evaluation of implicit shape estimation.

cs.CV↗

arXiv2021

Exploring intermediate representation for monocular vehicle pose estimation

We present a new learning-based framework to recover vehicle pose in SO(3) from a single RGB image. In contrast to previous works that map from local appearance to observation angles, we explore a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs) to estimate egocentric vehicle orientation. This approach features a deep model that transforms perceived intensities to IGRs, which are mapped to a 3D representation encoding object orientation in the camera coordinate system. Core problems are what IGRs to use and how to learn them more effectively. We answer the former question by designing IGRs based on an interpolated cuboid that derives from primitive 3D annotation readily. The latter question motivates us to incorporate geometry knowledge with a new loss function based on a projective invariant. This loss function allows unlabeled data to be used in the training stage to improve representation learning. Without additional labels, our system outperforms previous monocular RGB-based methods for joint vehicle detection and pose estimation on the KITTI benchmark, achieving performance even comparable to stereo methods. Code and pre-trained models are available at this https URL.

cs.CV↗

arXiv2021

Topological magnon insulator spin excitations in the two-dimensional ferromagnet CrBr$_3$

Topological magnons are bosonic analogues of topological fermions in electronic systems. They have been studied extensively by theory but rarely realized by experiment. Here, by performing inelastic neutron scattering measurements on single crystals of a two-dimensional ferromagnet CrBr$_3$, which was classified as Dirac magnon semimetal featured by the linear bands crossing at the Dirac points, we fully map out the magnetic excitation spectra, and reveal that there is an apparent gap of $\sim$3.5~meV between the acoustic and optical branches of the magnons at the K point. By collaborative efforts between experiment and theoretical calculations using a five-orbital Hubbard model obtained from first-principles calculations to derive the exchange parameters, we find that a Hamiltonian with Heisenberg exchange interactions, next-nearest-neighbor Dzyaloshinskii-Moriya (DM) interaction, and single-ion anisotropy is more appropriate to describe the system. Calculations using the model show that the lower and upper magnon bands separated by the gap exhibit Chern numbers of $\pm1$. These results indicate that CrBr$_3$ is a topological magnon insulator, where the nontrivial gap is a result of the DM interaction.

cond-mat.str-el↗

arXiv2021

How Do Adam and Training Strategies Help BNNs Optimization?

The best performing Binary Neural Networks (BNNs) are usually attained using Adam optimization and its multi-step training variants. However, to the best of our knowledge, few studies explore the fundamental reasons why Adam is superior to other optimizers like SGD for BNN optimization or provide analytical explanations that support specific training strategies. To address this, in this paper we first investigate the trajectories of gradients and weights in BNNs during the training process. We show the regularization effect of second-order momentum in Adam is crucial to revitalize the weights that are dead due to the activation saturation in BNNs. We find that Adam, through its adaptive learning rate strategy, is better equipped to handle the rugged loss surface of BNNs and reaches a better optimum with higher generalization ability. Furthermore, we inspect the intriguing role of the real-valued weights in binary networks, and reveal the effect of weight decay on the stability and sluggishness of BNN optimization. Through extensive experiments and analysis, we derive a simple training scheme, building on existing Adam-based optimization, which achieves 70.5% top-1 accuracy on the ImageNet dataset using the same architecture as the state-of-the-art ReActNet while achieving 1.1% higher accuracy. Code and models are available at https://github.com/liuzechun/AdamBNN.

cs.LG↗

arXiv2021

Cascaded deep monocular 3D human pose estimation with evolutionary training data

End-to-end deep representation learning has achieved remarkable accuracy for monocular 3D human pose estimation, yet these models may fail for unseen poses with limited and fixed training data. This paper proposes a novel data augmentation method that: (1) is scalable for synthesizing massive amount of training data (over 8 million valid 3D human poses with corresponding 2D projections) for training 2D-to-3D networks, (2) can effectively reduce dataset bias. Our method evolves a limited dataset to synthesize unseen 3D human skeletons based on a hierarchical human representation and heuristics inspired by prior knowledge. Extensive experiments show that our approach not only achieves state-of-the-art accuracy on the largest public benchmark, but also generalizes significantly better to unseen and rare poses. Code, pre-trained models and tools are available at this HTTPS URL.

cs.CV↗

arXiv2020

GSNet: Joint Vehicle Pose and Shape Reconstruction with Geometrical and Scene-aware Supervision

We present a novel end-to-end framework named as GSNet (Geometric and Scene-aware Network), which jointly estimates 6DoF poses and reconstructs detailed 3D car shapes from single urban street view. GSNet utilizes a unique four-way feature extraction and fusion scheme and directly regresses 6DoF poses and shapes in a single forward pass. Extensive experiments show that our diverse feature extraction and fusion scheme can greatly improve model performance. Based on a divide-and-conquer 3D shape representation strategy, GSNet reconstructs 3D vehicle shape with great detail (1352 vertices and 2700 faces). This dense mesh representation further leads us to consider geometrical consistency and scene context, and inspires a new multi-objective loss function to regularize network training, which in turn improves the accuracy of 6D pose estimation and validates the merit of jointly performing both tasks. We evaluate GSNet on the largest multi-task ApolloCar3D benchmark and achieve state-of-the-art performance both quantitatively and qualitatively. Project page is available at https://lkeab.github.io/gsnet/.

cs.CV↗