A5101790974- Topic Modeling
- Tactile and Sensory Interactions
- Interactive and Immersive Displays
- Natural Language Processing Techniques
- Multimodal Machine Learning Applications
- Modular Robots and Swarm Intelligence
- Augmented Reality Applications
- Advanced Sensor and Energy Harvesting Materials
- Domain Adaptation and Few-Shot Learning
- Semantic Web and Ontologies
- Advanced Image and Video Retrieval Techniques
- Biomedical Text Mining and Ontologies
- Innovative Human-Technology Interaction
- Metal-Organic Frameworks: Synthesis and Applications
- Text Readability and Simplification
- Data Quality and Management
- Gaze Tracking and Assistive Technology
- Electrowetting and Microfluidic Technologies
- PARP inhibition in cancer therapy
- Matrix Theory and Algorithms
- Diverse Topics in Contemporary Research
- Spinal Hematomas and Complications
- Human Pose and Action Recognition
- Insect and Arachnid Ecology and Behavior
- Sphingolipid Metabolism and Signaling
Tsinghua University
2020-2025
University of California, Berkeley
2024-2025
Huadong Hospital
2024
Fudan University
2024
Tianjin Hospital
2024
Tianjin University
2024
University Town of Shenzhen
2022-2024
Tianjin Medical University
2020-2023
Computer Network Information Center
2022
University of Chinese Academy of Sciences
2021-2022
Venous Materials is a novel concept and approach of an interactive material utilizing fluidic channels. We present design method for mechanisms that respond to deformation by mechanical inputs from the user, such as pressure bending. designed set primitive venous structures act embedded analog sensors, displaying flow color change. In this paper, we consider fluid medium drive tangible information triggered deformation, at same time, function responsive display information. To provide users...
We introduce Thermotion, a novel method using thermofluidic composites to design and display thermochromic animation effects on object surfaces. With fluidic channels embedded under the surfaces, utilize flows dynamically control surface temperature as an actuator for paints, which enables researchers designers first time create animations not only two three-dimensional surfaces but also made of few flexible everyday materials. report space with six primitives modification effects, we...
The key to preserving traditional crafts lies in living transmission, which is inseparable from sustaining artistic production, audience consumption, and progressive innovation with the physical media. As HCI researchers, we focus on hybrid field, involves numerous cross-disciplinary integration cases between craftsmanship digital technology at level, providing inspiration for innovating enlivening crafts. We conducted a multi-perspective review of 85 craft articles related over past decade,...
While the majority of pneumatic interfaces are powered and controlled by traditional electric pumps valves, alternative sustainable energy-harnessing technology has been attracting attention. This paper presents a novel solution to this challenge with development Sustainflatable system, self-sustaining system that can harvest renewable energy sources such as wind, water flow, moisture, sunlight, convert into compressed air, store it for later use in programmable intelligent way. The is...
Digital technologies can conduct an important role in preserving intangible cultural heritage (ICH). Nonetheless, existing work tends to be limited digital storage, presentation, dissemination, and education, with comparatively little concentrating on production reproduction of the craft, key revitalizing ICH. In this paper, we explore making as approach for both inheritance innovation ICH handicrafts. Taking Hairy Monkey craftsmanship instance, conducted a workshop 15 groups makers,...
Accurate and effective directional feedback is crucial for an electronic traveling aid device that guides visually impaired people in walking through paths. This paper presents Tactile Compass, a hand-held provides continuous with rotatable needle pointing toward the planned direction. We conducted two lab studies to evaluate effectiveness of solution. Results showed that, using participants could reach target direction place mean deviation 3.03° smoothly navigate along paths 60cm width,...
Plant life plays a critical role in the ecosystem. However, it is difficult for humans to perceive plants' reactions because biopotential and biochemical responses are invisible humans. Guttation droplets contain various chemicals which can reflect plant physiology environmental conditions real-time. Traditionally, these collected manually analyzed lab with expensive instruments. Here, we introduce Sensor, first on-site low-cost monitoring technology guttation droplets. This innovative...
Abstract As an age‐related disease, intervertebral disc degeneration is closely related to inflammation and aging. Inflammatory cytokines cellular senescence collectively contribute the degradation of disc. Blocking this synergy reduces extracellular matrix damage caused by In study, drug‐loaded nanofibers with sequential targeting functions are constructed through intelligent response, hydrophilicity, in situ self‐assembly empowerment flurbiprofen. The peptide precursor responds cleavage...
New materials and technologies facilitate the design of thermochromic dynamic paintings. However, creating a painting requires knowledge electrical engineering computer science, which is barrier for artists enthusiasts with non-technology backgrounds. Existing toolkits only support limited space fail to provide usable solutions independent creation meeting needs artists. We present ViviPaint, toolkit that assists in paintings easily conveniently. summarized pain points challenges by...
Large language models (LLMs) have shown impressive abilities for open-domain NLP tasks. However, LLMs are sometimes too footloose natural understanding (NLU) tasks which always restricted output and input format. Their performances on NLU highly related to prompts or demonstrations be poor at performing several representative tasks, such as event extraction entity typing. To this end, we present SeqGPT, a bilingual (i.e., English Chinese) open-source autoregressive model specially enhanced...
Dysregulation of SMAD4 (i.e. somatic mutation) is strongly associated with poor pancreatic ductal adenocarcinoma (PDAC) prognosis, yet the molecular mechanisms remain underlying this relationship obscure. Previously, we discovered that mutation renders cancer resistant to radiotherapy via promotion autophagy. In current work, observed a downregulation protein level in PDAC as compared adjacent normal tissue, and such
<title>Abstract</title> The recent surge of Multimodal Large Language Models (MLLMs) has fundamentally reshaped the landscape AI research and industry, shedding light on a promising path toward next milestone. However, significant challenges remain preventing MLLMs from being practical in real-world applications. most notable challenge comes huge cost running an MLLM with massive number parameters extensive computation. As result, need to be deployed high-performing cloud servers, which...
Entity Set Expansion (ESE) is a promising task which aims to expand entities of the target semantic class described by small seed entity set. Various NLP and IR applications will benefit from ESE due its ability discover knowledge. Although previous methods have achieved great progress, most them still lack handle hard negative (i.e., that are difficult distinguish entities), since two may or not belong same based on different granularity levels we analyze on. To address this challenge,...
The innovative millifluidic chamber-based pneumatic interface facilitates the creation of high-frequency, sequentially controlled, and finely detailed deformations—challenging to achieve using traditional setups. This advancement finds applications in information visualization, interactive entertainment, product design. Yet, designing such interfaces demands extensive empirical knowledge, predicting their deformations remains complex, requiring exhaustive experimentation design iterations....
The Entity Set Expansion (ESE) task aims to expand a handful of seed entities with new belonging the same semantic class. Conventional ESE methods are based on mono-modality (i.e., literal modality), which struggle deal complex in real world such as (1) Negative fine-grained differences. (2) Synonymous entities. (3) Polysemous (4) Long-tailed These challenges prompt us propose novel Multi-modal (MESE), where models integrate information from multiple modalities represent Intuitively,...
Phrase grounding aims to localize the objects described by phrases in a natural language specification. Previous works model interaction of inputs from text modality and visual only intra-modal global level consequently lacks ability capture precise complete context information. In this paper, we propose novel Cross-Modal Omni Interaction network (COI Net) composed neighboring module, cross-modal module multilevel alignment module. Our approach formulates complex spatial semantic...
Visual Relation Detection (VRD) aims to detect relationships between objects for image understanding. Most existing VRD methods rely on thousands of training samples each relationship achieve satisfactory performance. Some recent papers tackle this problem by few-shot learning with elaborately designed pipelines and pre-trained word vectors. However, the performance models is severely hampered poor generalization capability, as they struggle handle vast semantic diversity visual...
Large-scale commonsense knowledge bases empower a broad range of AI applications, where the automatic extraction (CKE) is fundamental and challenging problem. CKE from text known for suffering inherent sparsity reporting bias in text. Visual perception, on other hand, contains rich about real-world entities, e.g., (person, can_hold, bottle), which can serve as promising sources acquiring grounded knowledge. In this work, we present CLEVER, formulates distantly supervised multi-instance...
Thermal interfaces in HCI have been widely explored. Previous work either used electronics that continuously consume energy to actively control the temperatures, or relied on materials passively create only one predefined temperature behavior, which both leave gap of exploring objects allow self-contained programmable behaviors, utilizing materials' dynamic thermal properties.