arXiv subjects

Yueqi Li

Publications and source records attributed to Yueqi Li.

8 recordsLinked to original sources

arXiv2025

Foundation Models for Autonomous Driving Perception: A Survey Through Core Capabilities

Foundation models are revolutionizing autonomous driving perception, transitioning the field from narrow, task-specific deep learning models to versatile, general-purpose architectures trained on vast, diverse datasets. This survey examines how these models address critical challenges in autonomous perception, including limitations in generalization, scalability, and robustness to distributional shifts. The survey introduces a novel taxonomy structured around four essential capabilities for robust performance in dynamic driving environments: generalized knowledge, spatial understanding, multi-sensor robustness, and temporal reasoning. For each capability, the survey elucidates its significance and comprehensively reviews cutting-edge approaches. Diverging from traditional method-centric surveys, our unique framework prioritizes conceptual design principles, providing a capability-driven guide for model development and clearer insights into foundational aspects. We conclude by discussing key challenges, particularly those associated with the integration of these capabilities into real-time, scalable systems, and broader deployment challenges related to computational demands and ensuring model reliability against issues like hallucinations and out-of-distribution failures. The survey also outlines crucial future research directions to enable the safe and effective deployment of foundation models in autonomous driving systems.

cs.RO↗

arXiv2025

CX-Mind: A Pioneering Multimodal Large Language Model for Interleaved Reasoning in Chest X-ray via Curriculum-Guided Reinforcement Learning

Chest X-ray (CXR) imaging is one of the most widely used diagnostic modalities in clinical practice, encompassing a broad spectrum of diagnostic tasks. Recent advancements have seen the extensive application of reasoning-based multimodal large language models (MLLMs) in medical imaging to enhance diagnostic efficiency and interpretability. However, existing multimodal models predominantly rely on "one-time" diagnostic approaches, lacking verifiable supervision of the reasoning process. This leads to challenges in multi-task CXR diagnosis, including lengthy reasoning, sparse rewards, and frequent hallucinations. To address these issues, we propose CX-Mind, the first generative model to achieve interleaved "think-answer" reasoning for CXR tasks, driven by curriculum-based reinforcement learning and verifiable process rewards (CuRL-VPR). Specifically, we constructed an instruction-tuning dataset, CX-Set, comprising 708,473 images and 2,619,148 samples, and generated 42,828 high-quality interleaved reasoning data points supervised by clinical reports. Optimization was conducted in two stages under the Group Relative Policy Optimization framework: initially stabilizing basic reasoning with closed-domain tasks, followed by transfer to open-domain diagnostics, incorporating rule-based conditional process rewards to bypass the need for pretrained reward models. Extensive experimental results demonstrate that CX-Mind significantly outperforms existing medical and general-domain MLLMs in visual understanding, text generation, and spatiotemporal alignment, achieving an average performance improvement of 25.1% over comparable CXR-specific models. On real-world clinical dataset (Rui-CXR), CX-Mind achieves a mean recall@1 across 14 diseases that substantially surpasses the second-best results, with multi-center expert evaluations further confirming its clinical utility across multiple dimensions.

cs.LG↗

arXiv2023

Supervision Interpolation via LossMix: Generalizing Mixup for Object Detection and Beyond

The success of data mixing augmentations in image classification tasks has been well-received. However, these techniques cannot be readily applied to object detection due to challenges such as spatial misalignment, foreground/background distinction, and plurality of instances. To tackle these issues, we first introduce a novel conceptual framework called Supervision Interpolation (SI), which offers a fresh perspective on interpolation-based augmentations by relaxing and generalizing Mixup. Based on SI, we propose LossMix, a simple yet versatile and effective regularization that enhances the performance and robustness of object detectors and more. Our key insight is that we can effectively regularize the training on mixed data by interpolating their loss errors instead of ground truth labels. Empirical results on the PASCAL VOC and MS COCO datasets demonstrate that LossMix can consistently outperform state-of-the-art methods widely adopted for detection. Furthermore, by jointly leveraging LossMix with unsupervised domain adaptation, we successfully improve existing approaches and set a new state of the art for cross-domain object detection.

cs.CV↗

arXiv2023

Machine-Learning Prediction of the Computed Band Gaps of Double Perovskite Materials

Prediction of the electronic structure of functional materials is essential for the engineering of new devices. Conventional electronic structure prediction methods based on density functional theory (DFT) suffer from not only high computational cost, but also limited accuracy arising from the approximations of the exchange-correlation functional. Surrogate methods based on machine learning have garnered much attention as a viable alternative to bypass these limitations, especially in the prediction of solid-state band gaps, which motivated this research study. Herein, we construct a random forest regression model for band gaps of double perovskite materials, using a dataset of 1306 band gaps computed with the GLLBSC (Gritsenko, van Leeuwen, van Lenthe, and Baerends solid correlation) functional. Among the 20 physical features employed, we find that the bulk modulus, superconductivity temperature, and cation electronegativity exhibit the highest importance scores, consistent with the physics of the underlying electronic structure. Using the top 10 features, a model accuracy of 85.6% with a root mean square error of 0.64 eV is obtained, comparable to previous studies. Our results are significant in the sense that they attest to the potential of machine learning regressions for the rapid screening of promising candidate functional materials.

cond-mat.mtrl-sci↗

arXiv2022

Toward Edge-Efficient Dense Predictions with Synergistic Multi-Task Neural Architecture Search

In this work, we propose a novel and scalable solution to address the challenges of developing efficient dense predictions on edge platforms. Our first key insight is that MultiTask Learning (MTL) and hardware-aware Neural Architecture Search (NAS) can work in synergy to greatly benefit on-device Dense Predictions (DP). Empirical results reveal that the joint learning of the two paradigms is surprisingly effective at improving DP accuracy, achieving superior performance over both the transfer learning of single-task NAS and prior state-of-the-art approaches in MTL, all with just 1/10th of the computation. To the best of our knowledge, our framework, named EDNAS, is the first to successfully leverage the synergistic relationship of NAS and MTL for DP. Our second key insight is that the standard depth training for multi-task DP can cause significant instability and noise to MTL evaluation. Instead, we propose JAReD, an improved, easy-to-adopt Joint Absolute-Relative Depth loss, that reduces up to 88% of the undesired noise while simultaneously boosting accuracy. We conduct extensive evaluations on standard datasets, benchmark against strong baselines and state-of-the-art approaches, as well as provide an analysis of the discovered optimal architectures.

cs.CV↗

arXiv2022

Metal-insulator transition in the disordered Hubbard model of the Lieb lattice

Using the determinant quantum Monte Carlo method, we investigate the metal-insulator transition in the interacting disordered Hubbard model of a Lieb lattice, in which the system characterizes the flat band centered at the Fermi level. By choosing suitable electron densities, we ensure the weak interaction sign problem to improve the reliability of our results. It is found that disorder and on-site Coulomb repulsive interaction produce interesting effects that induce the metal-insulator transition which is impossible in the half-filled case. The density of states at the Fermi energy is still finite in the thermodynamic limit, suggesting that the system is an Anderson insulator rather than a Mott insulator. Moreover, this doping system is paramagnetic, unlike the half-filled system, which is ferrimagnetic.

cond-mat.str-el↗

arXiv2022

Doping-dependent metal-insulator transition in a disordered Hubbard model

We study the effect of disorder and doping on the metal-insulator transition in a repulsive Hubbard model on a square lattice using the determinant quantum Monte Carlo method. First, with the aim of making our results reliable, we compute the sign problem with various parameters such as temperature, disorder, on-site interactions, and lattice size. We show that in the presence of randomness in the hopping elements, the metal-insulator transition occurs and the critical disorder strength differs at different fillings. We also demonstrate that doping is a driving force behind the metal-insulator transition.

cond-mat.str-el↗

arXiv2019

Gate controlled quantum interference: direct observation of anti-resonances in single molecule charge transport

Quantum interference can profoundly affect charge transport in single molecules, but experiments can usually measure only the conductance at the Fermi energy. Because in general the most pronounced features of the quantum interference are not located at the Fermi energy, it is highly desirable to probe charge transport in a broader energy range. Here by the method of electrochemical gating, we measure the conductance and map the transmission functions of single molecules at and around the Fermi energy, and study signatures associated with constructive and destructive interference. With the electrochemical gate control, we tune the quantum interference between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), and directly observe anti-resonance, a distinct feature of destructive interference. By tuning the molecule in and out of anti-resonance, we achieve continuous control of the conductance over 2 orders of magnitude with a subthreshold swing of ~17 mV/dec, features relevant to high-speed and low-power electronics.

cond-mat.mes-hall↗