arXiv subjects

Zhanpeng Jin

Publications and source records attributed to Zhanpeng Jin.

At least 19 recordsLinked to original sources

arXiv2026

HAIGEN: Towards Human-AI Collaboration for Facilitating Creativity and Style Generation in Fashion Design

The process of fashion design usually involves sketching, refining, and coloring, with designers drawing inspiration from various images to fuel their creative endeavors. However, conventional image search methods often yield irrelevant results, impeding the design process. Moreover, creating and coloring sketches can be time-consuming and demanding, acting as a bottleneck in the design workflow. In this work, we introduce HAIGEN (Human-AI Collaboration for GENeration), an efficient fashion design system for Human-AI collaboration developed to aid designers. Specifically, HAIGEN consists of four modules. T2IM, located in the cloud, generates reference inspiration images directly from text prompts. With three other modules situated locally, the I2SM batch generates the image material library into a certain designer-style sketch material library. The SRM recommends similar sketches in the generated library to designers for further refinement, and the STM colors the refined sketch according to the styles of inspiration images. Through our system, any designer can perform local personalized fine-tuning and leverage the powerful generation capabilities of large models in the cloud, streamlining the entire design development process. Given that our approach integrates both cloud and local model deployment schemes, it effectively safeguards design privacy by avoiding the need to upload personalized data from local designers. We validated the effectiveness of each module through extensive qualitative and quantitative experiments. User surveys also confirmed that HAIGEN offers significant advantages in design efficiency, positioning it as a new generation of aid-tool for designers.

cs.HC↗

arXiv2026

TokenMem: Faithful Knowledge Injection for Frozen LLMs

Retrieval-augmented generation (RAG) enhances large language models (LLMs) with external knowledge, but suffers from knowledge conflicts: when retrieved information contradicts parametric memory, the shared self-attention pathway produces unpredictable outputs. We present TokenMem, a lightweight memory system that injects knowledge into frozen LLMs through a dedicated cross-attention channel, bypassing competition with parametric memory in the residual stream. TokenMem trains only a thin gating adapter ($\sim$3-7M parameters) via a two-phase curriculum: first learning general knowledge utilization, then strengthening faithful compliance under counterfactual knowledge. In controlled experiments on five models spanning three families (Qwen3-4B/8B/14B, LLaMA-3.1-8B, OLMo-3-7B), TokenMem achieves 69-70% Knowledge Compliance (KC) on counterfactual benchmarks, compared to 20-52% for vanilla RAG, a gap of up to 49 percentage points. Ablation studies show that the two-phase curriculum is critical: removing Phase 2 collapses KC to near-zero. Mechanistic analysis reveals that the gate adapter learns a conflict-aware, layer-specific injection strategy without explicit supervision.

cs.AI↗

arXiv2026

MyoSem: Aligning Electromyography to Natural-Language Action Semantics for Hand Action Understanding

Electromyography (EMG) directly reflects muscle activation and is a key sensing modality for gesture recognition, prosthetic control, and wearable interaction. Existing EMG methods, however, commonly formulate hand action understanding as classification over fixed labels, making it difficult to support querying, retrieval, and generalization based on action descriptions. We present MyoSem, an EMG--action semantic alignment framework that maps low-level EMG signals into a shared semantic space constructed from multi-view action descriptions. MyoSem combines multi-view action-semantic construction, activation-aware EMG encoding, and semantic query alignment, enabling bidirectional retrieval between EMG signals and text descriptions. We systematically evaluate MyoSem on EMG2Pose and NinaPro-series datasets. Results show that MyoSem performs well on EMG--text bidirectional retrieval, generally outperforms most baselines, and shows favorable generalization to unseen users, held-out action classes, and amputee-user transfer scenarios. Ablations and visualizations further validate the effectiveness of each module. Overall, MyoSem advances EMG-based hand action understanding from fixed-label recognition toward queryable bidirectional semantic retrieval, providing a new modeling paradigm for language-mediated EMG action understanding.

cs.CV↗

arXiv2026

MindMelody: A Closed-Loop EEG-Driven System for Personalized Music Intervention

Driven by the escalating global burden of mental health conditions, music-based interventions have attracted significant attention as a non-invasive, cost-effective modality for emotion regulation and psychological stress relief. However, current digital music services rely on static preferences and fail to adapt to users' instantaneous psychological states. Furthermore, directly mapping electroencephalography (EEG) to music generation remains challenging due to severe paired-data scarcity and a lack of interpretability. To address these limitations, we propose MindMelody, a fully functional, closed-loop real-time system for EEG-driven personalized music intervention. MindMelody introduces an emotion-mediated semantic bridge. Specifically, a hybrid Transformer-GNN first decodes real-time EEG signals into global Valence-Arousal states and local temporal affect trajectories. These states are then fed into a Retrieval-Augmented Generation (RAG)-equipped Large Language Model (LLM) to formulate structured intervention plans. Subsequently, a novel Hierarchical EEG Controller injects global affect prefixes and local temporal guidance into a pretrained music backbone, enabling fine-grained controllable audio synthesis. Crucially, the system incorporates a continuous feedback loop that updates generation parameters on the fly based on the user's evolving EEG dynamics. Extensive experiments show that MindMelody improves control adherence and emotional alignment, and receives higher perceived helpfulness in a short-term listening setting, suggesting its promise as an adaptive affect-aware music generation framework.

cs.SD↗

arXiv2026

EmoMM: Benchmarking and Steering MLLM for Multimodal Emotion Recognition under Conflict and Missingness

Multimodal Emotion Recognition (MER) is critical for interpreting real-world interactions. While Multimodal Large Language Models (MLLM) have shown promise in MER, their internal decision-making mechanisms under modality conflict and missingness remain largely underexplored. In this paper, to systematically investigate these behaviors, we introduce EmoMM, a comprehensive benchmark featuring modality-aligned, conflict, and missing subsets. Through extensive evaluation, we uncover a Video Contribution Collapse (VCC) phenomenon, where MLLM marginalize video evidence due to high token redundancy and modality preferences. To address this, we propose Conflict-aware Head-level Attention Steering (CHASE), a lightweight mechanism that detects modality conflicts and performs inference-time attention steering, effectively mitigating decision bias without retraining the backbone. Experimental results demonstrate that CHASE consistently improves performance across various settings, significantly enhancing the reliability of MLLM in complex affective scenarios.

cs.CV↗

arXiv2026

Persona-E$^2$: A Human-Grounded Dataset for Personality-Shaped Emotional Responses to Textual Events

Most affective computing research treats emotion as a static property of text, focusing on the writer's sentiment while overlooking the reader's perspective. This approach ignores how individual personalities lead to diverse emotional appraisals of the same event. Although role-playing Large Language Models (LLMs) attempt to simulate such nuanced reactions, they often suffer from "personality illusion'' -- relying on surface-level stereotypes rather than authentic cognitive logic. A critical bottleneck is the absence of ground-truth human data to link personality traits to emotional shifts. To bridge the gap, we introduce Persona-E$^2$ (Persona-Event2Emotion), a large-scale dataset grounded in annotated MBTI and Big Five traits to capture reader-based emotional variations across news, social media, and life narratives. Extensive experiments reveal that state-of-the-art LLMs struggle to capture precise appraisal shifts, particularly in social media domains. Crucially, we find that personality information significantly improves comprehension, with the Big Five traits alleviating "personality illusion.'

cs.CL↗

arXiv2026

AICA-Bench: Holistically Examining the Capabilities of VLMs in Affective Image Content Analysis

Vision-Language Models (VLMs) have demonstrated strong capabilities in perception, yet holistic Affective Image Content Analysis (AICA), which integrates perception, reasoning, and generation into a unified framework, remains underexplored. To address this gap, we introduce AICA-Bench, a comprehensive benchmark with three core tasks: Emotion Understanding (EU), Emotion Reasoning (ER), and Emotion-Guided Content Generation (EGCG). We evaluate 23 VLMs and identify two major limitations: weak intensity calibration and shallow open-ended descriptions. To address these issues, we propose Grounded Affective Tree (GAT) Prompting, a training-free framework that combines visual scaffolding with hierarchical reasoning. Experiments show that GAT reduces intensity errors and improves descriptive depth, providing a strong baseline for future research on affective multimodal understanding and generation.

cs.CV↗

arXiv2025

IMUFace: Towards Always-On 3D Facial Reconstruction via Earphone Inertial Sensing

The potential of facial expression reconstruction technology is significant, with applications in various fields such as human-computer interaction, affective computing, and virtual reality. Recent studies have proposed using ear-worn devices for facial expression reconstruction to address the environmental limitations and privacy concerns associated with traditional camera-based methods. However, these approaches still require improvements in terms of aesthetics and power consumption. This paper introduces a system called IMUFace. It uses inertial measurement units (IMUs) embedded in wireless earphones to detect subtle ear movements caused by facial muscle activities, allowing for covert and low-power facial reconstruction. A user study involving 12 participants was conducted, and a deep learning model named IMUTwinTrans was proposed. The results show that IMUFace can accurately predict users' facial landmarks with a precision of 2.21 mm, using only five minutes of training data. The predicted landmarks can be utilized to reconstruct a three-dimensional facial model. IMUFace operates at a sampling rate of 30 Hz with a relatively low power consumption of 58 mW. The findings presented in this study demonstrate the real-world applicability of IMUFace and highlight potential directions for further research to facilitate its practical adoption.

cs.HC↗

arXiv2025

FRAME: Feedback-Refined Agent Methodology for Enhancing Medical Research Insights

The automation of scientific research through large language models (LLMs) presents significant opportunities but faces critical challenges in knowledge synthesis and quality assurance. We introduce Feedback-Refined Agent Methodology (FRAME), a novel framework that enhances medical paper generation through iterative refinement and structured feedback. Our approach comprises three key innovations: (1) A structured dataset construction method that decomposes 4,287 medical papers into essential research components through iterative refinement; (2) A tripartite architecture integrating Generator, Evaluator, and Reflector agents that progressively improve content quality through metric-driven feedback; and (3) A comprehensive evaluation framework that combines statistical metrics with human-grounded benchmarks. Experimental results demonstrate FRAME's effectiveness, achieving significant improvements over conventional approaches across multiple models (9.91% average gain with DeepSeek V3, comparable improvements with GPT-4o Mini) and evaluation dimensions. Human evaluation confirms that FRAME-generated papers achieve quality comparable to human-authored works, with particular strength in synthesizing future research directions. The results demonstrated our work could efficiently assist medical research by building a robust foundation for automated medical research paper generation while maintaining rigorous academic standards.

cs.CL↗

arXiv2025

ExplicitLM: Decoupling Knowledge from Parameters via Explicit Memory Banks

Large language models suffer from knowledge staleness and lack of interpretability due to implicit knowledge storage across entangled network parameters, preventing targeted updates and reasoning transparency. We propose ExplicitLM, a novel architecture featuring a million-scale external memory bank storing human-readable knowledge as token sequences, enabling direct inspection and modification. We design a differentiable two-stage retrieval mechanism with efficient coarse-grained filtering via product key decomposition (reducing complexity from $\mathcal{O}(N \cdot |I|)$ to $\mathcal{O}(\sqrt{N} \cdot |I|)$) and fine-grained Gumbel-Softmax matching for end-to-end training. Inspired by dual-system cognitive theory, we partition knowledge into frozen explicit facts (20%) and learnable implicit patterns (80%), maintained through Exponential Moving Average updates for stability. ExplicitLM achieves up to 43.67% improvement on knowledge-intensive tasks versus standard Transformers, with 3.62$\times$ gains in low-data regimes (10k samples). Analysis shows strong correlations between memory retrieval and performance, with correct predictions achieving 49% higher hit rates. Unlike RAG systems with frozen retrieval, our jointly optimized architecture demonstrates that interpretable, updatable models can maintain competitive performance while providing unprecedented knowledge transparency.

cs.AI↗

arXiv2025

Wrist2Finger: Sensing Fingertip Force for Force-Aware Hand Interaction with a Ring-Watch Wearable

Hand pose tracking is essential for advancing applications in human-computer interaction. Current approaches, such as vision-based systems and wearable devices, face limitations in portability, usability, and practicality. We present a novel wearable system that reconstructs 3D hand pose and estimates per-finger forces using a minimal ring-watch sensor setup. A ring worn on the finger integrates an inertial measurement unit (IMU) to capture finger motion, while a smartwatch-based single-channel electromyography (EMG) sensor on the wrist detects muscle activations. By leveraging the complementary strengths of motion sensing and muscle signals, our approach achieves accurate hand pose tracking and grip force estimation in a compact wearable form factor. We develop a dual-branch transformer network that fuses IMU and EMG data with cross-modal attention to predict finger joint positions and forces simultaneously. A custom loss function imposes kinematic constraints for smooth force variation and realistic force saturation. Evaluation with 20 participants performing daily object interaction gestures demonstrates an average Mean Per Joint Position Error (MPJPE) of 0.57 cm and a fingertip force estimation (RMSE: 0.213, r=0.76). We showcase our system in a real-time Unity application, enabling virtual hand interactions that respond to user-applied forces. This minimal, force-aware tracking system has broad implications for VR/AR, assistive prosthetics, and ergonomic monitoring.

cs.HC↗

arXiv2025

Empathy-R1: A Chain-of-Empathy and Reinforcement Learning Framework for Long-Form Mental Health Support

Empathy is critical for effective mental health support, especially when addressing Long Counseling Texts (LCTs). However, existing Large Language Models (LLMs) often generate replies that are semantically fluent but lack the structured reasoning necessary for genuine psychological support, particularly in a Chinese context. To bridge this gap, we introduce Empathy-R1, a novel framework that integrates a Chain-of-Empathy (CoE) reasoning process with Reinforcement Learning (RL) to enhance response quality for LCTs. Inspired by cognitive-behavioral therapy, our CoE paradigm guides the model to sequentially reason about a help-seeker's emotions, causes, and intentions, making its thinking process both transparent and interpretable. Our framework is empowered by a new large-scale Chinese dataset, Empathy-QA, and a two-stage training process. First, Supervised Fine-Tuning instills the CoE's reasoning structure. Subsequently, RL, guided by a dedicated reward model, refines the therapeutic relevance and contextual appropriateness of the final responses. Experiments show that Empathy-R1 achieves strong performance on key automatic metrics. More importantly, human evaluations confirm its superiority, showing a clear preference over strong baselines and achieving a Win@1 rate of 44.30% on our new benchmark. By enabling interpretable and contextually nuanced responses, Empathy-R1 represents a significant advancement in developing responsible and genuinely beneficial AI for mental health support.

cs.CL↗

arXiv2025

Collaborative Medical Triage under Uncertainty: A Multi-Agent Dynamic Matching Approach

The post-pandemic surge in healthcare demand, coupled with critical nursing shortages, has placed unprecedented pressure on medical triage systems, necessitating innovative AI-driven solutions. We present a multi-agent interactive intelligent system for medical triage that addresses three fundamental challenges in current AI-based triage systems: inadequate medical specialization leading to misclassification, heterogeneous department structures across healthcare institutions, and inefficient detail-oriented questioning that impedes rapid triage decisions. Our system employs three specialized agents--RecipientAgent, InquirerAgent, and DepartmentAgent--that collaborate through Inquiry Guidance mechanism and Classification Guidance Mechanism to transform unstructured patient symptoms into accurate department recommendations. To ensure robust evaluation, we constructed a comprehensive Chinese medical triage dataset from "Ai Ai Yi Medical Network", comprising 3,360 real-world cases spanning 9 primary departments and 62 secondary departments. Experimental results demonstrate that our multi-agent system achieves 89.6% accuracy in primary department classification and 74.3% accuracy in secondary department classification after four rounds of patient interaction. The system's dynamic matching based guidance mechanisms enable efficient adaptation to diverse hospital configurations while maintaining high triage accuracy. We successfully developed this multi-agent triage system that not only adapts to organizational heterogeneity across healthcare institutions but also ensures clinically sound decision-making.

cs.AI↗

arXiv2025

xIDS-EnsembleGuard: An Explainable Ensemble Learning-based Intrusion Detection System

In this paper, we focus on addressing the challenges of detecting malicious attacks in networks by designing an advanced Explainable Intrusion Detection System (xIDS). The existing machine learning and deep learning approaches have invisible limitations, such as potential biases in predictions, a lack of interpretability, and the risk of overfitting to training data. These issues can create doubt about their usefulness, transparency, and a decrease in trust among stakeholders. To overcome these challenges, we propose an ensemble learning technique called "EnsembleGuard." This approach uses the predicted outputs of multiple models, including tree-based methods (LightGBM, GBM, Bagging, XGBoost, CatBoost) and deep learning models such as LSTM (long short-term memory) and GRU (gated recurrent unit), to maintain a balance and achieve trustworthy results. Our work is unique because it combines both tree-based and deep learning models to design an interpretable and explainable meta-model through model distillation. By considering the predictions of all individual models, our meta-model effectively addresses key challenges and ensures both explainable and reliable results. We evaluate our model using well-known datasets, including UNSW-NB15, NSL-KDD, and CIC-IDS-2017, to assess its reliability against various types of attacks. During analysis, we found that our model outperforms both tree-based models and other comparative approaches in different attack scenarios.

cs.CR↗

arXiv2023

5G/6G-Enabled Metaverse Technologies: Taxonomy, Applications, and Open Security Challenges with Future Research Directions

Internet technology has proven to be a vital contributor to many cutting-edge innovations that have given humans access to interact virtually with objects. Until now, numerous virtual systems had been developed for digital transformation to enable access to thousands of services and applications that range from virtual gaming to social networks. However, the majority of these systems lack to maintain consistency during interconnectivity and communication. To explore this discussion, in the recent past a new term, Metaverse has been introduced, which is the combination of meta and universe that describes a shared virtual environment, where a number of technologies, such as 4th and 5th generation technologies, VR, ML algorithms etc., work collectively to support each other for the sake of one objective, which is the virtual accessibility of objects via one network platform. With the development, integration, and virtualization of technologies, a lot of improvement in daily life applications is expected, but at the same time, there is a big challenge for the research community to secure this platform from external and external threats, because this technology is exposed to many cybersecurity attacks. Hence, it is imperative to systematically review and understand the taxonomy, applications, open security challenges, and future research directions of the emerging Metaverse technologies. In this paper, we have made useful efforts to present a comprehensive survey regarding Metaverse technology by taking into account the aforesaid parameters. Following this, in the initial phase, we explored the future of Metaverse in the presence of 4th and 5th generation technologies. Thereafter, we discussed the possible attacks to set a preface for the open security challenges. Based on that, we suggested potential research directions that could be beneficial to address these challenges cost-effectively.

cs.CR↗

arXiv2022

EEGLog: Lifelogging EEG Data When You Listen to Music

Self-tracking has been long discussed, which can monitor daily activities and help users to recall previous experiences. Such data-capturing technique is no longer limited to photos, text messages, or personal diaries in recent years. With the development of wearable EEG devices, we introduce a novel modality of logging EEG data while listening to music, and bring up the idea of the neural-centric way of life with the designed data analysis application named EEGLog. Four consumer-grade wearable EEG devices are explored by collecting EEG data from 24 participants. Three modules are introduced in EEGLog, including the summary module of EEG data, emotion reports, music listening activities, and memorial moments, the emotion detection module, and the music recommendation module. Feedback from interviews about using EEG devices and EEGLog were obtained and analyzed for future EEG logging development.

cs.HC↗

arXiv2022

BEAMERS: Brain-Engaged, Active Music-based Emotion Regulation System

With the increasing demands of emotion comprehension and regulation in our daily life, a customized music-based emotion regulation system is introduced by employing current EEG information and song features, which predicts users' emotion variation in the valence-arousal model before recommending music. The work shows that: (1) a novel music-based emotion regulation system with a commercial EEG device is designed without employing deterministic emotion recognition models for daily usage; (2) the system considers users' variant emotions towards the same song, and by which calculate user's emotion instability and it is in accordance with Big Five Personality Test; (3) the system supports different emotion regulation styles with users' designation of desired emotion variation, and achieves an accuracy of over $0.85$ with 2-seconds EEG data; (4) people feel easier to report their emotion variation comparing with absolute emotional states, and would accept a more delicate music recommendation system for emotion regulation according to the questionnaire.

cs.HC↗

arXiv2022

STT: Soft Template Tuning for Few-Shot Adaptation

Prompt tuning has been an extremely effective tool to adapt a pre-trained model to downstream tasks. However, standard prompt-based methods mainly consider the case of sufficient data of downstream tasks. It is still unclear whether the advantage can be transferred to the few-shot regime, where only limited data are available for each downstream task. Although some works have demonstrated the potential of prompt-tuning under the few-shot setting, the main stream methods via searching discrete prompts or tuning soft prompts with limited data are still very challenging. Through extensive empirical studies, we find that there is still a gap between prompt tuning and fully fine-tuning for few-shot learning. To bridge the gap, we propose a new prompt-tuning framework, called Soft Template Tuning (STT). STT combines manual and auto prompts, and treats downstream classification tasks as a masked language modeling task. Comprehensive evaluation on different settings suggests STT can close the gap between fine-tuning and prompt-based methods without introducing additional parameters. Significantly, it can even outperform the time- and resource-consuming fine-tuning method on sentiment classification tasks.

cs.CL↗