arXiv subjects

Zecheng Li

Publications and source records attributed to Zecheng Li.

17 recordsLinked to original sources

arXiv2026

Depth-Semantic Alignment and Affinity-Guided Fusion for Structured Radar Point Cloud Generation

Point clouds are an important carrier of three-dimensional spatial information, and their quality directly affects the performance of downstream perception tasks such as object detection and tracking. However, millimeter-wave radar point clouds are typically sparse, noisy, and structurally incomplete. To address these limitations, this paper proposes a multimodal point cloud generation method based on vision-radar fusion. The proposed method leverages image semantic information to impose structural constraints and achieve spatial alignment for radar point clouds, while incorporating a sparse completion strategy to enhance point density and recover missing structures. The generated point clouds are further evaluated in object detection and tracking tasks. Experimental results demonstrate that the proposed method effectively improves point cloud quality and enhances the detection accuracy and robustness of perception models in complex environments, providing a practical solution for multisensor point cloud generation and intelligent perception systems.

cs.CV↗

arXiv2026

Thouless Pumping of Large Chern Numbers in Optical Floquet Quasicrystals

Chern numbers are central to correlated and topological phenomena, yet most topological systems are associated with Chern numbers of order unity. Here we propose a scheme to achieve large Chern numbers in an optical Floquet quasicrystal with cold atoms, which can be directly measured via Thouless pumping. We study the quasienergy spectrum of Floquet quasicrystals and characterize the emergent Chern numbers using gap labeling theorem. We further investigate the Thouless pumping in the Floquet quasicrystal at different driving frequencies and amplitudes, revealing the connection between transport features and the quasienergy spectrum. Our findings open new avenues for exploring rich topological dynamics in Floquet quasicrystals and realizing fractional Chern insulating states.

cond-mat.quant-gas↗

arXiv2026

Internalizing Meta-Experience into Memory for Guided Reinforcement Learning in Large Language Models

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an effective approach for enhancing the reasoning capabilities of Large Language Models (LLMs). Despite its efficacy, RLVR faces a meta-learning bottleneck: it lacks mechanisms for error attribution and experience internalization intrinsic to the human learning cycle beyond practice and verification, thereby limiting fine-grained credit assignment and reusable knowledge formation. We term such reusable knowledge representations derived from past errors as meta-experience. Based on this insight, we propose Meta-Experience Learning (MEL), a novel framework that incorporates self-distilled meta-experience into the model's parametric memory. Building upon standard RLVR, we introduce an additional design that leverages the LLM's self-verification capability to conduct contrastive analysis on paired correct and incorrect trajectories, identify the precise bifurcation points where reasoning errors arise, and summarize them into generalizable meta-experience. The meta-experience is further internalized into the LLM's parametric memory by minimizing the negative log-likelihood, which induces a language-modeled reward signal that bridges correct and incorrect reasoning trajectories and facilitates effective knowledge reuse. Experimental results demonstrate that MEL achieves consistent improvements on benchmarks, yielding 3.92%--4.73% Pass@1 gains across varying model sizes.

cs.LG↗

arXiv2026

MagicGUI-RMS: A Multi-Agent Reward Model System for Self-Evolving GUI Agents via Automated Feedback Reflux

Graphical user interface (GUI) agents are rapidly progressing toward autonomous interaction and reliable task execution across diverse applications. However, two central challenges remain unresolved: automating the evaluation of agent trajectories and generating high-quality training data at scale to enable continual improvement. Existing approaches often depend on manual annotation or static rule-based verification, which restricts scalability and limits adaptability in dynamic environments. We present MagicGUI-RMS, a multi-agent reward model system that delivers adaptive trajectory evaluation, corrective feedback, and self-evolving learning capabilities. MagicGUI-RMS integrates a Domain-Specific Reward Model (DS-RM) with a General-Purpose Reward Model (GP-RM), enabling fine-grained action assessment and robust generalization across heterogeneous GUI tasks. To support reward learning at scale, we design a structured data construction pipeline that automatically produces balanced and diverse reward datasets, effectively reducing annotation costs while maintaining sample fidelity. During execution, the reward model system identifies erroneous actions, proposes refined alternatives, and continuously enhances agent behavior through an automated data-reflux mechanism. Extensive experiments demonstrate that MagicGUI-RMS yields substantial gains in task accuracy, behavioral robustness. These results establish MagicGUI-RMS as a principled and effective foundation for building self-improving GUI agents driven by reward-based adaptation.

cs.AI↗

arXiv2026

From Evidence-Based Medicine to Knowledge Graph: Retrieval-Augmented Generation for Sports Rehabilitation and a Domain Benchmark

Current medical retrieval-augmented generation (RAG) approaches overlook evidence-based medicine (EBM) principles, leading to two key gaps: (1) the lack of PICO alignment between queries and retrieved evidence, and (2) the absence of evidence hierarchy considerations during reranking. We present SR-RAG, an EBM-adapted GraphRAG framework that integrates the PICO framework into knowledge graph construction and retrieval, and proposes Bayesian Evidence Tier Reranking (BETR) to calibrate ranking scores by evidence grade without predefined weights. Validated in sports rehabilitation, we release a knowledge graph (357,844 nodes, 371,226 edges) and a benchmark of 1,637 QA pairs. SR-RAG achieves 0.812 evidence recall@10, 0.830 nugget coverage, 0.819 answer faithfulness, 0.882 semantic similarity, and 0.788 PICOT match accuracy, substantially outperforming five baselines. Five expert clinicians rated the system 4.66--4.84 on a 5-point Likert scale, and system rankings are preserved on a human-verified gold subset (n=80).

cs.CL↗

arXiv2025

SEASONED: Semantic-Enhanced Self-Counterfactual Explainable Detection of Adversarial Exploiter Contracts

Decentralized Finance (DeFi) attacks have resulted in significant losses, often orchestrated through Adversarial Exploiter Contracts (AECs) that exploit vulnerabilities in victim smart contracts. To proactively identify such threats, this paper targets the explainable detection of AECs. Existing detection methods struggle to capture semantic dependencies and lack interpretability, limiting their effectiveness and leaving critical knowledge gaps in AEC analysis. To address these challenges, we introduce SEASONED, an effective, self-explanatory, and robust framework for AEC detection. SEASONED extracts semantic information from contract bytecode to construct a semantic relation graph (SRG), and employs a self-counterfactual explainable detector (SCFED) to classify SRGs and generate explanations that highlight the core attack logic. SCFED further enhances robustness, generalizability, and data efficiency by extracting representative information from these explanations. Both theoretical analysis and experimental results demonstrate the effectiveness of SEASONED, which showcases outstanding detection performance, robustness, generalizability, and data efficiency learning ability. To support further research, we also release a new dataset of 359 AECs.

cs.CR↗

arXiv2025

SLRTP2025 Sign Language Production Challenge: Methodology, Results, and Future Work

Sign Language Production (SLP) is the task of generating sign language video from spoken language inputs. The field has seen a range of innovations over the last few years, with the introduction of deep learning-based approaches providing significant improvements in the realism and naturalness of generated outputs. However, the lack of standardized evaluation metrics for SLP approaches hampers meaningful comparisons across different systems. To address this, we introduce the first Sign Language Production Challenge, held as part of the third SLRTP Workshop at CVPR 2025. The competition's aims are to evaluate architectures that translate from spoken language sentences to a sequence of skeleton poses, known as Text-to-Pose (T2P) translation, over a range of metrics. For our evaluation data, we use the RWTH-PHOENIX-Weather-2014T dataset, a German Sign Language - Deutsche Gebardensprache (DGS) weather broadcast dataset. In addition, we curate a custom hidden test set from a similar domain of discourse. This paper presents the challenge design and the winning methodologies. The challenge attracted 33 participants who submitted 231 solutions, with the top-performing team achieving BLEU-1 scores of 31.40 and DTW-MJE of 0.0574. The winning approach utilized a retrieval-based framework and a pre-trained language model. As part of the workshop, we release a standardized evaluation network, including high-quality skeleton extraction-based keypoints establishing a consistent baseline for the SLP field, which will enable future researchers to compare their work against a broader range of methods.

cs.CV↗

arXiv2025

Time-Dependent Hamiltonian Simulation via Time-Independent Dynamics in a Larger Space

In this paper, we present a proof-of-concept quantum algorithm for simulating time-dependent Hamiltonian evolution by reducing the problem to simulating a time-independent Hamiltonian in a larger space using a discrete clock Hamiltonian construction. A similar construction was first explored for this simulation problem by Watkins, Wiebe, Roggero, and Lee [PRX Quantum, 2024]. Our algorithm improves upon their work in terms of the dependence on evolution time and precision. In addition, the complexity matches the state-of-the-art simulation algorithms using other approaches. To achieve this improvement, we use Duhamel's principle to treat the clock and system Hamiltonians separately and exploit properties of Gaussian quadrature to reduce the simulation cost. Our approach demonstrates that time-dependent Hamiltonian simulation can be as efficient in a simpler framework and hence provides a new angle to model and simulate time-dependent systems.

quant-ph↗

arXiv2025

CUTHERMO: Understanding GPU Memory Inefficiencies with Heat Map Profiling

GPUs have become indispensable in high-performance computing, machine learning, and many other domains. Efficiently utilizing the memory subsystem on GPUs is critical for maximizing computing power through massive parallelism. Analyzing memory access patterns has proven to be an effective method for understanding memory bottlenecks in applications. However, comprehensive runtime and fine-grained memory profiling support is lacking on GPU architectures. In this work, we introduce cuThermo, a lightweight and practical profiling tool for GPU memory analysis. It operates on GPU binaries without requiring any modifications to hardware, operating system, or application source code. Given a CUDA application, cuThermo identifies memory inefficiencies at runtime via a heat map based on distinct visited warp counts to represent word-sector-level data sharing and provides optimization guidance in performance tuning iterations. Through our experiments on six applications, we identified five memory access patterns that are portable across different GPU architectures. By evaluating optimization on two GPUs, cuThermo achieves up to $721.79\%$ performance improvement.

cs.DC↗

arXiv2025

VisualPuzzles: Decoupling Multimodal Reasoning Evaluation from Domain Knowledge

Current multimodal benchmarks often conflate reasoning with domain-specific knowledge, making it difficult to isolate and evaluate general reasoning abilities in non-expert settings. To address this, we introduce VisualPuzzles, a benchmark that targets visual reasoning while deliberately minimizing reliance on specialized knowledge. VisualPuzzles consists of diverse questions spanning five categories: algorithmic, analogical, deductive, inductive, and spatial reasoning. One major source of our questions is manually translated logical reasoning questions from the Chinese Civil Service Examination. Experiments show that VisualPuzzles requires significantly less intensive domain-specific knowledge and more complex reasoning compared to benchmarks like MMMU, enabling us to better evaluate genuine multimodal reasoning. Evaluations show that state-of-the-art multimodal large language models consistently lag behind human performance on VisualPuzzles, and that strong performance on knowledge-intensive benchmarks does not necessarily translate to success on reasoning-focused, knowledge-light tasks. Additionally, reasoning enhancements such as scaling up inference compute (with "thinking" modes) yield inconsistent gains across models and task types, and we observe no clear correlation between model size and performance. We also found that models exhibit different reasoning and answering patterns on VisualPuzzles compared to benchmarks with heavier emphasis on knowledge. VisualPuzzles offers a clearer lens through which to evaluate reasoning capabilities beyond factual recall and domain knowledge.

cs.CL↗

arXiv2025

Cross-Modal Consistency Learning for Sign Language Recognition

Pre-training has been proven to be effective in boosting the performance of Isolated Sign Language Recognition (ISLR). Existing pre-training methods solely focus on the compact pose data, which eliminates background perturbation but inevitably suffers from insufficient semantic cues compared to raw RGB videos. Nevertheless, learning representation directly from RGB videos remains challenging due to the presence of sign-independent visual features. To address this dilemma, we propose a Cross-modal Consistency Learning framework (CCL-SLR), which leverages the cross-modal consistency from both RGB and pose modalities based on self-supervised pre-training. First, CCL-SLR employs contrastive learning for instance discrimination within and across modalities. Through the single-modal and cross-modal contrastive learning, CCL-SLR gradually aligns the feature spaces of RGB and pose modalities, thereby extracting consistent sign representations. Second, we further introduce Motion-Preserving Masking (MPM) and Semantic Positive Mining (SPM) techniques to improve cross-modal consistency from the perspective of data augmentation and sample similarity, respectively. Extensive experiments on four ISLR benchmarks show that CCL-SLR achieves impressive performance, demonstrating its effectiveness. The code will be released to the public.

cs.CV↗

arXiv2025

Uni-Sign: Toward Unified Sign Language Understanding at Scale

Sign language pre-training has gained increasing attention for its ability to enhance performance across various sign language understanding (SLU) tasks. However, existing methods often suffer from a gap between pre-training and fine-tuning, leading to suboptimal results. To address this, we propose Uni-Sign, a unified pre-training framework that eliminates the gap between pre-training and downstream SLU tasks through a large-scale generative pre-training strategy and a novel fine-tuning paradigm. First, we introduce CSL-News, a large-scale Chinese Sign Language (CSL) dataset containing 1,985 hours of video paired with textual annotations, which enables effective large-scale pre-training. Second, Uni-Sign unifies SLU tasks by treating downstream tasks as a single sign language translation (SLT) task during fine-tuning, ensuring seamless knowledge transfer between pre-training and fine-tuning. Furthermore, we incorporate a prior-guided fusion (PGF) module and a score-aware sampling strategy to efficiently fuse pose and RGB information, addressing keypoint inaccuracies and improving computational efficiency. Extensive experiments across multiple SLU benchmarks demonstrate that Uni-Sign achieves state-of-the-art performance across multiple downstream SLU tasks. Dataset and code are available at github.com/ZechengLi19/Uni-Sign.

cs.CV↗

arXiv2024

Scaling up Multimodal Pre-training for Sign Language Understanding

Sign language serves as the primary meaning of communication for the deaf-mute community. Different from spoken language, it commonly conveys information by the collaboration of manual features, i.e., hand gestures and body movements, and non-manual features, i.e., facial expressions and mouth cues. To facilitate communication between the deaf-mute and hearing people, a series of sign language understanding (SLU) tasks have been studied in recent years, including isolated/continuous sign language recognition (ISLR/CSLR), gloss-free sign language translation (GF-SLT) and sign language retrieval (SL-RT). Sign language recognition and translation aims to understand the semantic meaning conveyed by sign languages from gloss-level and sentence-level, respectively. In contrast, SL-RT focuses on retrieving sign videos or corresponding texts from a closed-set under the query-by-example search paradigm. These tasks investigate sign language topics from diverse perspectives and raise challenges in learning effective representation of sign language videos. To advance the development of sign language understanding, exploring a generalized model that is applicable across various SLU tasks is a profound research direction.

cs.CV↗

arXiv2024

SEDS: Semantically Enhanced Dual-Stream Encoder for Sign Language Retrieval

Different from traditional video retrieval, sign language retrieval is more biased towards understanding the semantic information of human actions contained in video clips. Previous works typically only encode RGB videos to obtain high-level semantic features, resulting in local action details drowned in a large amount of visual information redundancy. Furthermore, existing RGB-based sign retrieval works suffer from the huge memory cost of dense visual data embedding in end-to-end training, and adopt offline RGB encoder instead, leading to suboptimal feature representation. To address these issues, we propose a novel sign language representation framework called Semantically Enhanced Dual-Stream Encoder (SEDS), which integrates Pose and RGB modalities to represent the local and global information of sign language videos. Specifically, the Pose encoder embeds the coordinates of keypoints corresponding to human joints, effectively capturing detailed action features. For better context-aware fusion of two video modalities, we propose a Cross Gloss Attention Fusion (CGAF) module to aggregate the adjacent clip features with similar semantic information from intra-modality and inter-modality. Moreover, a Pose-RGB Fine-grained Matching Objective is developed to enhance the aggregated fusion feature by contextual matching of fine-grained dual-stream features. Besides the offline RGB encoder, the whole framework only contains learnable lightweight networks, which can be trained end-to-end. Extensive experiments demonstrate that our framework significantly outperforms state-of-the-art methods on various datasets.

cs.CV↗

arXiv2024

Efficient Optimal Control of Open Quantum Systems

The optimal control problem for open quantum systems can be formulated as a time-dependent Lindbladian that is parameterized by a number of time-dependent control variables. Given an observable and an initial state, the goal is to tune the control variables so that the expected value of some observable with respect to the final state is maximized. In this paper, we present algorithms for solving this optimal control problem efficiently, i.e., having a poly-logarithmic dependency on the system dimension, which is exponentially faster than best-known classical algorithms. Our algorithms are hybrid, consisting of both quantum and classical components. The quantum procedure simulates time-dependent Lindblad evolution that drives the initial state to the final state, and it also provides access to the gradients of the objective function via quantum gradient estimation. The classical procedure uses the gradient information to update the control variables. At the technical level, we provide the first (to the best of our knowledge) simulation algorithm for time-dependent Lindbladians with an $\ell_1$-norm dependence. As an alternative, we also present a simulation algorithm in the interaction picture to improve the algorithm for the cases where the time-independent component of a Lindbladian dominates the time-dependent part. On the classical side, we heavily adapt the state-of-the-art classical optimization analysis to interface with the quantum part of our algorithms. Both the quantum simulation techniques and the classical optimization analyses might be of independent interest.

quant-ph↗

arXiv2024

Complete Instances Mining for Weakly Supervised Instance Segmentation

Weakly supervised instance segmentation (WSIS) using only image-level labels is a challenging task due to the difficulty of aligning coarse annotations with the finer task. However, with the advancement of deep neural networks (DNNs), WSIS has garnered significant attention. Following a proposal-based paradigm, we encounter a redundant segmentation problem resulting from a single instance being represented by multiple proposals. For example, we feed a picture of a dog and proposals into the network and expect to output only one proposal containing a dog, but the network outputs multiple proposals. To address this problem, we propose a novel approach for WSIS that focuses on the online refinement of complete instances through the use of MaskIoU heads to predict the integrity scores of proposals and a Complete Instances Mining (CIM) strategy to explicitly model the redundant segmentation problem and generate refined pseudo labels. Our approach allows the network to become aware of multiple instances and complete instances, and we further improve its robustness through the incorporation of an Anti-noise strategy. Empirical evaluations on the PASCAL VOC 2012 and MS COCO datasets demonstrate that our method achieves state-of-the-art performance with a notable margin. Our implementation will be made available at https://github.com/ZechengLi19/CIM.

cs.CV↗

arXiv2022

MDS-Net: A Multi-scale Depth Stratification Based Monocular 3D Object Detection Algorithm

Monocular 3D object detection is very challenging in autonomous driving due to the lack of depth information. This paper proposes a one-stage monocular 3D object detection algorithm based on multi-scale depth stratification, which uses the anchor-free method to detect 3D objects in a per-pixel prediction. In the proposed MDS-Net, a novel depth-based stratification structure is developed to improve the network's ability of depth prediction by establishing mathematical models between depth and image size of objects. A new angle loss function is then developed to further improve the accuracy of the angle prediction and increase the convergence speed of training. An optimized soft-NMS is finally applied in the post-processing stage to adjust the confidence of candidate boxes. Experiments on the KITTI benchmark show that the MDS-Net outperforms the existing monocular 3D detection methods in 3D detection and BEV detection tasks while fulfilling real-time requirements.

cs.CV↗