arXiv subjects

Yibo Lyu

Publications and source records attributed to Yibo Lyu.

9 recordsLinked to original sources

arXiv2026

EpiLENS: Patient-Relative Epileptogenic Zone Localization from Multi-Center Intracranial EEG

Drug-resistant epilepsy remains a major clinical challenge, as successful neurosurgery depends critically on accurate epileptogenic zone (EZ) localization accounting for substantial variability across patients, seizures, implantation layouts, recording systems and clinical centers. Existing intracranial electroencephalogram (iEEG) methods typically rely on channel-wise classifiers trained globally, which often obscure patient-specific electrophysiological abnormalities and exhibit instability under severe class imbalance and noisy clinical annotations. To address these limitations, we present EpiLENS, a primary-guided asymmetric dual-branch framework for patient-relative epileptogenic localization. Its localization strategy, Conservative Dual-Evidence Localization (CDEL), asymmetrically combines independently trained but complementary branches at inference: the Patient-Relative Quantile Network (PRQ-Net), which captures seizure-consistent deviations from each patient's internal electrophysiological baseline, and the Boundary-Coverage Ranking Network (BCR-Net), which emphasizes ambiguous epileptogenic zone or non-epileptogenic zone boundaries and recovery of the patient-specific epileptogenic set. CDEL retains PRQ-Net as the primary branch while incorporating complementary patient-wise ranking evidence from BCR-Net. Experiments on a heterogeneous four-center cohort demonstrate improved balanced localization over classical feature-based and raw-iEEG neural baselines, while component ablations and within-patient permutation analyses support the contributions of patient-relative normalization, lower-tail seizure aggregation, and boundary-coverage ranking. Cross-seizure and leave-one-center-out generalization experiments further confirm that the proposed patient-relative evidence remains robust across repeated recordings and transfers effectively to unseen clinical centers.

cs.CE↗

arXiv2026

Fuzzy-Geometric Branch-Point Modeling for Structure-Aware Augmentation of Handwritten Chinese Characters

Data scarcity and structural distortion significantly limit handwriting recognition in high-security authentication. Existing augmentation methods often cause topological and morphological damage, particularly when processing complex Chinese characters where stroke intersections, ligatures, and sharp turns render traditional branch-point detection unreliable. To address this, this paper proposes a fuzzy geometry-driven structure-aware (FGSA) augmentation framework. We model branch points as fuzzy sets within the skeleton space, constructing a continuous branch-point membership field by integrating topological neighborhood evidence with direction field divergence. This membership field is adaptively optimized via an unsupervised surrogate objective, enabling robust stroke decoupling without manual annotation. Finally, kinematically-aligned samples are synthesized through parameterized cubic B\'ezier reconstruction and multi-strategy perturbations, ensuring a balance between structural fidelity and sample diversity. Moreover, we establish LZUSig, a large-scale, highly challenging dataset specifically dedicated to fine-grained structural degradation in Chinese handwritten signatures. Extensive experiments on CASIA-HWDB1.1, ChiSig, and LZUSig demonstrate that FGSA significantly reduces the word-level error rate ($\Delta$WER), achieving optimal recognition gains over the compared baselines. More importantly, it strikes a robust trade-off among task gain, structural fidelity, and discriminative feature preservation, offering a highly controllable solution for handwriting augmentation.

cs.CV↗

arXiv2026

Enhancing the Code Reasoning Capabilities of LLMs via Consistency-based Reinforcement Learning

Code reasoning refers to the task of predicting the output of a program given its source code and specific inputs. It can measure the reasoning capability of large language models (LLMs) and also benefit downstream tasks such as code generation and mathematical reasoning. Existing work has verified the effectiveness of reinforcement learning on the task. However, these methods design rewards solely based on final outputs or coarse-grained signals, and neglect the inherent consistency of the stepwise reasoning process in the task. Therefore, these methods often result in sparse reward or reward hacking, which limits the full play of enhanced learning capabilities. To alleviate these issues, we propose CodeThinker, a consistency-driven reinforcement learning framework for code reasoning. Specifically, CodeThinker has three key components: (1) a stepwise reasoning-aware model training module, which utilizes a consistency tracing paradigm as a template to synthesize training data that captures the stepwise reasoning process; (2) a dynamic beam sampling strategy, which aims to improve the quality of sampled outputs under a fixed sampling budget; and (3) a consistency reward mechanism that can effectively alleviate reward hacking. Experiments on three popular benchmarks show that CodeThinker achieves state-of-the-art performance across multiple LLMs. For instance, it outperforms the strongest baseline by 4.3% in accuracy when deployed on Qwen2.5-Coder-7B-Instruct. We also validate the effectiveness of CodeThinker on downstream tasks. Results show that, without additional training, CodeThinker obtains average accuracy gains of 5.33 and 3.11 percentage points on mathematical reasoning and code reasoning tasks covering 17 programming languages, respectively.

cs.LG↗

arXiv2026

PersonalAlign: Hierarchical Implicit Intent Alignment for Personalized GUI Agent with Long-Term User-Centric Records

While GUI agents have shown strong performance under explicit and completion instructions, real-world deployment requires aligning with users' more complex implicit intents. In this work, we highlight Hierarchical Implicit Intent Alignment for Personalized GUI Agent (PersonalAlign), a new agent task that requires agents to leverage long-term user records as persistent context to resolve omitted preferences in vague instructions and anticipate latent routines by user state for proactive assistance. To facilitate this study, we introduce AndroidIntent, a benchmark designed to evaluate agents' ability in resolving vague instructions and providing proactive suggestions through reasoning over long-term user records. We annotated 775 user-specific preferences and 215 routines from 20k long-term records across different users for evaluation. Furthermore, we introduce Hierarchical Intent Memory Agent (HIM-Agent), which maintains a continuously updating personal memory and hierarchically organizes user preferences and routines for personalization. Finally, we evaluate a range of GUI agents on AndroidIntent, including GPT-5, Qwen3-VL, and UI-TARS, further results show that HIM-Agent significantly improves both execution and proactive performance by 15.7% and 7.3%.

cs.AI↗

arXiv2025

PUMA: Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning

As multimedia content expands, the demand for unified multimodal retrieval (UMR) in real-world applications increases. Recent work leverages multimodal large language models (MLLMs) to tackle this task. However, their large parameter size results in high training costs and low inference efficiency. To address this, we propose PUMA: a Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning. Our approach improves UMR from both structural and learning perspectives. (1) Structurally, we propose Layer-Pruned Self-Distillation, which prunes MLLMs by keeping only shallow layers while distilling features from dropped deep layers as teacher signals. This reduces parameters and preserves representation capability. (2) On the learning side, we introduce Modality-Adaptive Contrastive Learning Loss (MAC-Loss), which separates in-batch negatives into harder intra-modality and easier inter-modality groups based on the target modality, assigning different temperature strategies to enhance learning efficiency. Experiments show our method significantly reduces resource usage while maintaining strong performance.

cs.MM↗

arXiv2025

Less is More: Empowering GUI Agent with Context-Aware Simplification

The research focus of GUI agents is shifting from text-dependent to pure-vision-based approaches, which, though promising, prioritize comprehensive pre-training data collection while neglecting contextual modeling challenges. We probe the characteristics of element and history contextual modeling in GUI agent and summarize: 1) the high-density and loose-relation of element context highlight the existence of many unrelated elements and their negative influence; 2) the high redundancy of history context reveals the inefficient history modeling in current GUI agents. In this work, we propose a context-aware simplification framework for building an efficient and effective GUI Agent, termed SimpAgent. To mitigate potential interference from numerous unrelated elements, we introduce a masking-based element pruning method that circumvents the intractable relation modeling through an efficient masking mechanism. To reduce the redundancy in historical information, we devise a consistency-guided history compression module, which enhances implicit LLM-based compression through innovative explicit guidance, achieving an optimal balance between performance and efficiency. With the above components, SimpAgent reduces 27% FLOPs and achieves superior GUI navigation performances. Comprehensive navigation experiments across diverse web and mobile environments demonstrate the effectiveness and potential of our agent.

cs.CV↗

arXiv2024

Token-level Correlation-guided Compression for Efficient Multimodal Document Understanding

Cropping high-resolution document images into multiple sub-images is the most widely used approach for current Multimodal Large Language Models (MLLMs) to do document understanding. Most of current document understanding methods preserve all tokens within sub-images and treat them equally. This neglects their different informativeness and leads to a significant increase in the number of image tokens. To perform a more adaptive and efficient document understanding, we propose Token-level Correlation-guided Compression, a parameter-free and plug-and-play methodology to optimize token processing. Firstly, we propose an innovative approach for assessing the pattern repetitiveness based on the correlation between each patch tokens. This method identifies redundant tokens, allowing for the determination of the sub-image's information density. Secondly, we present a token-level sampling method that efficiently captures the most informative tokens by delving into the correlation between the [CLS] token and patch tokens. By integrating these strategies, we develop a plug-and-play adaptive compressor module that can be seamlessly incorporated into MLLMs utilizing cropping techniques. This module not only enhances the processing speed during training and inference but also maintains comparable performance. We conduct experiments with the SOTA document understanding model mPLUG-DocOwl1.5 and the effectiveness is demonstrated through extensive comparisons with other compression methods.

cs.CV↗

arXiv2021

A Scalable 256-Elements E-Band Phased-Array Transceiver for Broadband Communication

For E-band wireless communications, a high gain steerable antenna with sub-arrays is desired to reduce the implementation complexity. This paper presents an E-band communication link with 256-elements antennas based on 8-elements sub-arrays and four beam-forming chips in silicon germanium (SiGe) bipolar complementary metal-oxide-semiconductor (BiCMOS), which is packaged on a 19-layer low temperature co-fired ceramic (LTCC) substrate. After the design and manufacture of the 256-elements antenna, a fast near-field calibration method is proposed for calibration, where a single near-field measurement is required. Then near-field to far-field (NFFF) transform and far-field to near-field (FFNF) transform are used for the bore-sight calibration. The comparison with high frequency structure simulator (HFSS) is utilized for the non-bore-sight calibration. Verified on the 256-elements antenna, the beam-forming performance measured in the chamber is in good agreement with the simulations. The communication in the office environment is also realized using a fifth generation (5G) new radio (NR) system, whose bandwidth is 400 megahertz (MHz) and waveform format is orthogonal frequency division multiplexing (OFDM) with 120 kilohertz (kHz) sub-carrier spacing.

eess.SP↗

arXiv2020

Two Types of Mixed Orthogonal Frequency Division Multiplexing (X-OFDM) Waveform for Optical Wireless Communication

The optical wireless communication (OWC) with the intensity modulation (IM), requires the modulated radio frequency (RF) signal to be real and non-negative. To satisfy the requirements, this paper proposes two types of mixed orthogonal frequency division multiplexing (X-OFDM) waveform. The Hermitian symmetry (HS) character of the sub-carriers in the frequency domain, guarantees the signal in the time domain to be real, which reduces the spectral efficiency to $1/2$. For the odd sub-carriers in the frequency domain, the signal in the time domain after the inverse fast fourier transform (IFFT) is antisymmetric. For the even sub-carriers in the frequency domain, the signal in the time domain after the IFFT is symmetric. Based on the antisymmetric and symmetric characters, the two types of X-OFDM waveform are designed to guarantee the signal in the time domain to be non-negative, where the direct current (DC) bias is not needed. With $N$ sub-carriers in the frequency domain, the generated signal in the time domain has $3N/2$ points, which further reduces the spectral efficiency to $1/3 = 1/2 \times 2/3$. The numerical simulations show that, the two types of X-OFDM waveform greatly enhance the power efficiency considering the OWC channel with the signal-dependent noise and/or the signal-independent noise.

eess.SP↗