arXiv subjects

Han Chen

Publications and source records attributed to Han Chen.

At least 37 records · Page 2Linked to original sources

arXiv2026

SunnyParking: Multi-Shot Trajectory Generation and Motion State Awareness for Human-like Parking

Autonomous parking fundamentally differs from on-road driving due to its frequent direction changes and complex maneuvering requirements. However, existing End-to-End (E2E) planning methods often simplify the parking task into a geometric path regression problem, neglecting explicit modeling of the vehicle's kinematic state. This "dimensionality deficiency" easily leads to physically infeasible trajectories and deviates from real human driving behavior, particularly at critical gear-shift points in multi-shot parking scenarios. In this paper, we propose SunnyParking, a novel dual-branch E2E architecture that achieves motion state awareness by jointly predicting spatial trajectories and discrete motion state sequences (e.g., forward/reverse). Additionally, we introduce a Fourier feature-based representation of target parking slots to overcome the resolution limitations of traditional bird's-eye view (BEV) approaches, enabling high-precision target interactions. Experimental results demonstrate that our framework generates more robust and human-like trajectories in complex multi-shot parking scenarios, while significantly improving gear-shift point localization accuracy compared to state-of-the-art methods. We open-source a new parking dataset of the CARLA simulator, specifically designed to evaluate full prediction capabilities under complex maneuvers.

cs.RO↗

arXiv2026

Latent-Space Contrastive Reinforcement Learning for Stable and Efficient LLM Reasoning

While Large Language Models (LLMs) demonstrate exceptional performance in surface-level text generation, their nature in handling complex multi-step reasoning tasks often remains one of ``statistical fitting'' rather than systematic logical deduction. Traditional Reinforcement Learning (RL) attempts to mitigate this by introducing a ``think-before-speak'' paradigm. However, applying RL directly in high-dimensional, discrete token spaces faces three inherent challenges: sample-inefficient rollouts, high gradient estimation variance, and the risk of catastrophic forgetting. To fundamentally address these structural bottlenecks, we propose \textbf{DeepLatent Reasoning (DLR)}, a latent-space bidirectional contrastive reinforcement learning framework. This framework shifts the trial-and-error cost from expensive token-level full sequence generation to the continuous latent manifold. Specifically, we introduce a lightweight assistant model to efficiently sample $K$ reasoning chain encodings within the latent space. These encodings are filtered via a dual reward mechanism based on correctness and formatting; only high-value latent trajectories are fed into a \textbf{frozen main model} for single-pass decoding. To maximize reasoning diversity while maintaining coherence, we design a contrastive learning objective to enable directed exploration within the latent space. Since the main model parameters remain frozen during optimization, this method mathematically eliminates catastrophic forgetting. Experiments demonstrate that under comparable GPU computational budgets, DLR achieves more stable training convergence, supports longer-horizon reasoning chains, and facilitates the sustainable accumulation of reasoning capabilities, providing a viable path toward reliable and scalable reinforcement learning for LLMs.

cs.LG↗

arXiv2025

STARS: Semantic Tokens with Augmented Representations for Recommendation at Scale

Real-world ecommerce recommender systems must deliver relevant items under strict tens-of-milliseconds latency constraints despite challenges such as cold-start products, rapidly shifting user intent, and dynamic context including seasonality, holidays, and promotions. We introduce STARS, a transformer-based sequential recommendation framework built for large-scale, low-latency ecommerce settings. STARS combines several innovations: dual-memory user embeddings that separate long-term preferences from short-term session intent; semantic item tokens that fuse pretrained text embeddings, learnable deltas, and LLM-derived attribute tags, strengthening content-based matching, long-tail coverage, and cold-start performance; context-aware scoring with learned calendar and event offsets; and a latency-conscious two-stage retrieval pipeline that performs offline embedding generation and online maximum inner-product search with filtering, enabling tens-of-milliseconds response times. In offline evaluations on production-scale data, STARS improves Hit@5 by more than 75 percent relative to our existing LambdaMART system. A large-scale A/B test on 6 million visits shows statistically significant lifts, including Total Orders +0.8%, Add-to-Cart on Home +2.0%, and Visits per User +0.5%. These results demonstrate that combining semantic enrichment, multi-intent modeling, and deployment-oriented design can yield state-of-the-art recommendation quality in real-world environments without sacrificing serving efficiency.

cs.IR↗

arXiv2025

Causal Inference on Sequential Treatments via Tensor Completion

Marginal Structural Models (MSMs) are popular for causal inference of sequential treatments in longitudinal observational studies, which however are sensitive to model misspecification. To achieve flexible modeling, we envision the potential outcomes to form a three-dimensional tensor indexed by subject, time, and treatment regime and propose a tensorized history-restricted MSM (HRMSM). The semi-parametric tensor factor model allows us to leverage the underlying low-rank structure of the potential outcomes tensor and exploit the pre-treatment covariate information to recover the counterfactual outcomes. We incorporate the inverse probability of treatment weighting in the loss function for tensor completion to adjust for time-varying confounding. Theoretically, a non-asymptotic upper bound on the Frobenius norm error for the proposed estimator is provided. Empirically, simulation studies show that the proposed tensor completion approach outperforms the parametric HRMSM and existing matrix/tensor completion methods. Finally, we illustrate the practical utility of the proposed approach to study the effect of ventilation on organ dysfunction from the Medical Information Mart for Intensive Care database.

stat.ME↗

arXiv2025

Sliced Wasserstein Regression

While statistical modeling of distributional data has gained increased attention, the case of multivariate distributions has been somewhat neglected despite its relevance in various applications. This is because the Wasserstein distance, commonly used in distributional data analysis, poses challenges for multivariate distributions. A promising alternative is the sliced Wasserstein distance, which offers a computationally simpler solution. We propose distributional regression models with multivariate distributions as responses paired with Euclidean vector predictors. The foundation of our methodology is a slicing transform from the multivariate distribution space to the sliced distribution space for which we establish a theoretical framework, with the Radon transform as a prominent example. We introduce and study the asymptotic properties of sample-based estimators for two regression approaches, one based on utilizing the sliced Wasserstein distance directly in the multivariate distribution space, and a second approach based on a new slice-wise distance, employing a univariate distribution regression for each slice. Both global and local Fréchet regression methods are deployed for these approaches and illustrated in simulations and through applications. These include joint distributions of excess winter death rates and winter temperature anomalies in European countries as a function of base winter temperature and also data from finance.

stat.ME↗

arXiv2025

R-Capsule: Compressing High-Level Plans for Efficient Large Language Model Reasoning

Chain-of-Thought (CoT) prompting helps Large Language Models (LLMs) tackle complex reasoning by eliciting explicit step-by-step rationales. However, CoT's verbosity increases latency and memory usage and may propagate early errors across long chains. We propose the Reasoning Capsule (R-Capsule), a framework that aims to combine the efficiency of latent reasoning with the transparency of explicit CoT. The core idea is to compress the high-level plan into a small set of learned latent tokens (a Reasoning Capsule) while keeping execution steps lightweight or explicit. This hybrid approach is inspired by the Information Bottleneck (IB) principle, where we encourage the capsule to be approximately minimal yet sufficient for the task. Minimality is encouraged via a low-capacity bottleneck, which helps improve efficiency. Sufficiency is encouraged via a dual objective: a primary task loss for answer accuracy and an auxiliary plan-reconstruction loss that encourages the capsule to faithfully represent the original textual plan. The reconstruction objective helps ground the latent space, thereby improving interpretability and reducing the use of uninformative shortcuts. Our framework strikes a balance between efficiency, accuracy, and interpretability, thereby reducing the visible token footprint of reasoning while maintaining or improving accuracy on complex benchmarks. Our codes are available at: https://anonymous.4open.science/r/Reasoning-Capsule-7BE0

cs.CL↗

arXiv2025

Feather-SQL: A Lightweight NL2SQL Framework with Dual-Model Collaboration Paradigm for Small Language Models

Natural Language to SQL (NL2SQL) has seen significant advancements with large language models (LLMs). However, these models often depend on closed-source systems and high computational resources, posing challenges in data privacy and deployment. In contrast, small language models (SLMs) struggle with NL2SQL tasks, exhibiting poor performance and incompatibility with existing frameworks. To address these issues, we introduce Feather-SQL, a new lightweight framework tailored for SLMs. Feather-SQL improves SQL executability and accuracy through 1) schema pruning and linking, 2) multi-path and multi-candidate generation. Additionally, we introduce the 1+1 Model Collaboration Paradigm, which pairs a strong general-purpose chat model with a fine-tuned SQL specialist, combining strong analytical reasoning with high-precision SQL generation. Experimental results on BIRD demonstrate that Feather-SQL improves NL2SQL performance on SLMs, with around 10% boost for models without fine-tuning. The proposed paradigm raises the accuracy ceiling of SLMs to 54.76%, highlighting its effectiveness.

cs.CL↗

arXiv2025

The Manipulative Power of Voice Characteristics: Investigating Deceptive Patterns in Mandarin Chinese Female Synthetic Speech

Pervasive voice interaction enables deceptive patterns through subtle voice characteristics, yet empirical investigation into this manipulation lags behind, especially within major non-English language contexts. Addressing this gap, our study presents the first systematic investigation into voice characteristic-based dark patterns employing female synthetic voices in Mandarin Chinese. This focus is crucial given the prevalence of female personas in commercial assistants and the prosodic significance in the Chinese language. Guided by the conceptual framework identifying key influencing factors, we systematically evaluate effectiveness variations by manipulating voice characteristics (five characteristics, three intensities) across different scenarios (shopping vs. question-answering) with different commercial aims. A preliminary study (N=24) validated the experimental materials and the main study (N=36) revealed significant behavioral manipulation (up to +2027.6%). Crucially, the analysis showed that effectiveness varied significantly with voice characteristics and scenario, mediated by user perception (of tone, intonation, timbre) and user demographics (individual preferences, though limited demographic impact). These interconnected findings offer evidence-based insights for ethical design.

cs.HC↗

arXiv2025

Ferroelectricity in 6 Angstrom-Thick Two-dimensional Ga$_2$O$_3$

Atomic-scale ferroelectric thin films hold great promise for high-density, low-power applications but face stability and voltage scaling challenges at extreme thinness. Here, we demonstrate ferroelectricity in single-crystalline two-dimensional (2D) Ga$_2$O$_3$, an ultra-wide-bandgap semiconductor, at just 6 angstrom thickness, exhibiting exceptional retention and thermal stability. We show that epitaxial beta-Ga$_2$O$_3$ can be exfoliated down to a half-unit cell thickness via a self-limiting mechanism, enabling a biaxial strain-induced phase transition into a novel ferroelectric layered structure. Strain modulation enables the reduction of polarization switching voltage to 0.8 V, meeting CMOS voltage scaling requirements. Theoretical calculations reveal that switching is driven by covalent bond reconstruction, effectively countering depolarization and enhancing stability. Additionally, we integrate ferroelectric 2D Ga$_2$O$_3$ onto silicon using a low-temperature, back-end-of-line-compatible process. This work advances the exploration of sub-nanometer ferroelectrics, paving the way for high-density, low-power, non-volatile applications seamlessly integrated with advanced silicon technology.

cond-mat.mtrl-sci↗

arXiv2025

HEPP: Hyper-efficient Perception and Planning for High-speed Obstacle Avoidance of UAVs

High-speed obstacle avoidance of uncrewed aerial vehicles (UAVs) in cluttered environments is a significant challenge. Existing UAV planning and obstacle avoidance systems can only fly at moderate speeds or at high speeds over empty or sparse fields. In this article, we propose a hyper-efficient perception and planning system for the high-speed obstacle avoidance of UAVs. The system mainly consists of three modules: 1) A novel incremental robocentric mapping method with distance and gradient information, which takes 89.5% less time compared to existing methods. 2) A novel obstacle-aware topological path search method that generates multiple distinct paths. 3) An adaptive gradient-based high-speed trajectory generation method with a novel time pre-allocation algorithm. With these innovations, the system has an excellent real-time performance with only milliseconds latency in each iteration, taking 79.24% less time than existing methods at high speeds (15 m/s in cluttered environments), allowing UAVs to fly swiftly and avoid obstacles in cluttered environments. The planned trajectory of the UAV is close to the global optimum in both temporal and spatial domains. Finally, extensive validations in both simulation and real-world experiments demonstrate the effectiveness of our proposed system for high-speed navigation in cluttered environments.

cs.RO↗

arXiv2025

Enhancing breast cancer detection on screening mammogram using self-supervised learning and a hybrid deep model of Swin Transformer and Convolutional Neural Network

Purpose: The scarcity of high-quality curated labeled medical training data remains one of the major limitations in applying artificial intelligence (AI) systems to breast cancer diagnosis. Deep models for mammogram analysis and mass (or micro-calcification) detection require training with a large volume of labeled images, which are often expensive and time-consuming to collect. To reduce this challenge, we proposed a novel method that leverages self-supervised learning (SSL) and a deep hybrid model, named \textbf{HybMNet}, which combines local self-attention and fine-grained feature extraction to enhance breast cancer detection on screening mammograms. Approach: Our method employs a two-stage learning process: (1) SSL Pretraining: We utilize EsViT, a SSL technique, to pretrain a Swin Transformer (Swin-T) using a limited set of mammograms. The pretrained Swin-T then serves as the backbone for the downstream task. (2) Downstream Training: The proposed HybMNet combines the Swin-T backbone with a CNN-based network and a novel fusion strategy. The Swin-T employs local self-attention to identify informative patch regions from the high-resolution mammogram, while the CNN-based network extracts fine-grained local features from the selected patches. A fusion module then integrates global and local information from both networks to generate robust predictions. The HybMNet is trained end-to-end, with the loss function combining the outputs of the Swin-T and CNN modules to optimize feature extraction and classification performance. Results: The proposed method was evaluated for its ability to detect breast cancer by distinguishing between benign (normal) and malignant mammograms. Leveraging SSL pretraining and the HybMNet model, it achieved AUC of 0.864 (95% CI: 0.852, 0.875) on the CMMD dataset and 0.889 (95% CI: 0.875, 0.903) on the INbreast dataset, highlighting its effectiveness.

cs.CV↗

arXiv2025

Breast Cancer Detection from Multi-View Screening Mammograms with Visual Prompt Tuning

Accurate detection of breast cancer from high-resolution mammograms is crucial for early diagnosis and effective treatment planning. Previous studies have shown the potential of using single-view mammograms for breast cancer detection. However, incorporating multi-view data can provide more comprehensive insights. Multi-view classification, especially in medical imaging, presents unique challenges, particularly when dealing with large-scale, high-resolution data. In this work, we propose a novel Multi-view Visual Prompt Tuning Network (MVPT-NET) for analyzing multiple screening mammograms. We first pretrain a robust single-view classification model on high-resolution mammograms and then innovatively adapt multi-view feature learning into a task-specific prompt tuning process. This technique selectively tunes a minimal set of trainable parameters (7\%) while retaining the robustness of the pre-trained single-view model, enabling efficient integration of multi-view data without the need for aggressive downsampling. Our approach offers an efficient alternative to traditional feature fusion methods, providing a more robust, scalable, and efficient solution for high-resolution mammogram analysis. Experimental results on a large multi-institution dataset demonstrate that our method outperforms conventional approaches while maintaining detection efficiency, achieving an AUROC of 0.852 for distinguishing between Benign, DCIS, and Invasive classes. This work highlights the potential of MVPT-NET for medical imaging tasks and provides a scalable solution for integrating multi-view data in breast cancer detection.

cs.CV↗

arXiv2025

LogQuant: Log-Distributed 2-Bit Quantization of KV Cache with Superior Accuracy Preservation

We introduce LogQuant, a groundbreaking 2-bit quantization technique for KV Cache in large language model (LLM) inference, delivering substantial memory savings while preserving superior performance. Previous methods either assume that later tokens are more important or attempt to predict important tokens based on earlier attention patterns. Both approaches, however, can result in performance bottlenecks or frequent mispredictions. LogQuant takes a different approach. By applying a log-based filtering mechanism, it selectively compresses the KV Cache across the entire context, achieving better performance with the same or even reduced memory footprint compared to existing methods. In benchmark tests, it enhances throughput by 25% and boosts batch size by 60% without increasing memory consumption. For challenging tasks such as Math and Code Completion, LogQuant improves accuracy by 40% to 200% at the same compression ratio, outperforming comparable techniques.LogQuant integrates effortlessly with popular inference frameworks like Python's transformers library. Implementation can be available in https://github.com/Concyclics/LogQuantKV.

cs.LG↗

arXiv2025

Forming Auxiliary High-confident Instance-level Loss to Promote Learning from Label Proportions

Learning from label proportions (LLP), i.e., a challenging weakly-supervised learning task, aims to train a classifier by using bags of instances and the proportions of classes within bags, rather than annotated labels for each instance. Beyond the traditional bag-level loss, the mainstream methodology of LLP is to incorporate an auxiliary instance-level loss with pseudo-labels formed by predictions. Unfortunately, we empirically observed that the pseudo-labels are are often inaccurate due to over-smoothing, especially for the scenarios with large bag sizes, hurting the classifier induction. To alleviate this problem, we suggest a novel LLP method, namely Learning from Label Proportions with Auxiliary High-confident Instance-level Loss (L^2P-AHIL). Specifically, we propose a dual entropy-based weight (DEW) method to adaptively measure the confidences of pseudo-labels. It simultaneously emphasizes accurate predictions at the bag level and avoids overly smoothed predictions. We then form high-confident instance-level loss with DEW, and jointly optimize it with the bag-level loss in a self-training manner. The experimental results on benchmark datasets show that L^2P-AHIL can surpass the existing baseline methods, and the performance gain can be more significant as the bag size increases. The implementation of our method is available at https://github.com/TianhaoMa5/LLP-AHIL.

cs.AI↗

arXiv2025

RuleR: Improving LLM Controllability by Rule-based Data Recycling

Large language models (LLMs) still lack delicate controllability over their responses, which is critical to enhancing their performance and the user experience. However, curating supervised fine-tuning (SFT) datasets to improve LLM controllability usually relies on human experts or proprietary LLMs, which requires additional costs. To bridge this gap, we propose Rule-based Data Recycling (RuleR), a data augmentation method incorporating multiple constraints into the original data samples according to predefined rules, which creates new training tasks to consolidate the controllability of LLMs. Instead of creating new data from scratch, RuleR "recycles" existing data by simply applying rule-based edits to their responses and appending the rule-instructions in their original instructions. Experimental results demonstrate RuleR's effectiveness in improving LLM controllability while maintaining general instruction-following capabilities.

cs.CL↗

arXiv2025

Wafer-scale Integration of Single-Crystalline MoS$_2$ for Flexible Electronics Enabled by Oxide Dry-transfer

Atomically thin, single-crystalline transition metal dichalcogenides (TMDCs) grown via chemical vapor deposition (CVD) on sapphire substrates exhibit exceptional mechanical and electrical properties, positioning them as excellent channel materials for flexible electronics. However, conventional wet-transfer processes for integrating these materials onto flexible substrates often introduce surface contamination, significantly degrading device performance. Here, we present a wafer-scale dry-transfer technique using a high-dielectric oxide as the transfer medium, enabling the integration of 4-inch single-crystalline MoS$_2$ onto flexible substrates. This method eliminates contact with polymers or solvents, thus preserving the intrinsic electronic properties of MoS$_2$. As a result, the fabricated flexible field-effect transistor (FET) arrays exhibit remarkable performance, with a mobility of 117 cm$^2$/Vs, a subthreshold swing of 68.8 mV dec$^{-1}$, and an ultra-high current on/off ratio of $10^{12}$-values comparable to those achieved on rigid substrates. Leveraging the outstanding electrical characteristics, we demonstrated MoS$_2$-based flexible inverters operating in the subthreshold regime, achieving both a high gain of 218 and ultra-low power consumption of 1.4 pW/$μ$m. Additionally, we integrated a flexible tactile sensing system driven by active-matrix MoS$_2$ FET arrays onto a robotic gripper, enabling real-time object identification. These findings demonstrate the simultaneous achievement of high electrical performance and flexibility, highlighting the immense potential of single-crystalline TMDC-based flexible electronics for real-world applications.

physics.app-ph↗

arXiv2025

A VM-HDL Co-Simulation Framework for Systems with PCIe-Connected FPGAs

PCIe-connected FPGAs are gaining popularity as an accelerator technology in data centers. However, it is challenging to jointly develop and debug host software and FPGA hardware. Changes to the hardware design require a time-consuming FPGA synthesis process, and modification to the software, especially the operating system and device drivers, can frequently cause the system to hang, without providing enough information for debugging. The combination of these problems results in long debug iterations and a slow development process. To overcome these problems, we designed a VM-HDL co-simulation framework, which is capable of running the same software, operating system, and hardware designs as the target physical system, while providing full visibility and significantly shorter debug iterations.

cs.DC↗

arXiv2024

GenTel-Safe: A Unified Benchmark and Shielding Framework for Defending Against Prompt Injection Attacks

Large Language Models (LLMs) like GPT-4, LLaMA, and Qwen have demonstrated remarkable success across a wide range of applications. However, these models remain inherently vulnerable to prompt injection attacks, which can bypass existing safety mechanisms, highlighting the urgent need for more robust attack detection methods and comprehensive evaluation benchmarks. To address these challenges, we introduce GenTel-Safe, a unified framework that includes a novel prompt injection attack detection method, GenTel-Shield, along with a comprehensive evaluation benchmark, GenTel-Bench, which compromises 84812 prompt injection attacks, spanning 3 major categories and 28 security scenarios. To prove the effectiveness of GenTel-Shield, we evaluate it together with vanilla safety guardrails against the GenTel-Bench dataset. Empirically, GenTel-Shield can achieve state-of-the-art attack detection success rates, which reveals the critical weakness of existing safeguarding techniques against harmful prompts. For reproducibility, we have made the code and benchmarking dataset available on the project page at https://gentellab.github.io/gentel-safe.github.io/.

cs.CR↗