A5089885910- Particle physics theoretical and experimental studies
- High-Energy Particle Collisions Research
- Quantum Chromodynamics and Particle Interactions
- Particle Detector Development and Performance
- Dark Matter and Cosmic Phenomena
- Computational Physics and Python Applications
- Nuclear Physics and Applications
- Wind and Air Flow Studies
- Cosmology and Gravitation Theories
- Atomic and Subatomic Physics Research
- Multi-Agent Systems and Negotiation
- Radiation Detection and Scintillator Technologies
- Urologic and reproductive health conditions
- Climate variability and models
- Neutrino Physics Research
- Fish Ecology and Management Studies
- Astrophysics and Cosmic Phenomena
- Structural Health Monitoring Techniques
- Fuzzy Logic and Control Systems
- Meteorological Phenomena and Simulations
- Pharmacy and Medical Practices
- Seismic Performance and Analysis
- Infrastructure Maintenance and Monitoring
- Structural Response to Dynamic Loads
- Natural Language Processing Techniques
Heilongjiang Earthquake Agency
2025
Shanghai Astronomical Observatory
2025
University of Michigan
2015-2023
Université Savoie Mont Blanc
2023
Institut National de Physique Nucléaire et de Physique des Particules
2023
Nanjing University
2023
Institute of High Energy Physics
2003-2020
The University of Adelaide
2016
The Abdus Salam International Centre for Theoretical Physics (ICTP)
2011-2016
Istituto Nazionale di Fisica Nucleare, Gruppo Collegato di Udine
2011-2016
The scientific research paradigm is undergoing a profound transformation owing to the development of Artificial Intelligence (AI). Recent works demonstrate that various AI-assisted methods can largely improve efficiency by improving data analysis, accelerating computation, and fostering novel idea generation. To further move towards ultimate goal (i.e., automatic research), in this paper, we propose Dolphin, first closed-loop open-ended auto-research framework build entire process human...
ABSTRACT A room, as a critical unit of building, plays key role in fulfilling functional demands, which relies on the coordinated performance essential equipment and facilities. Understanding seismic damage mechanisms correlations among room components is crucial for assessing post‐earthquake functionality rooms entire building. In this study, shaking table tests full‐scale medical model were conducted to investigate these issues. The results revealed within same that patterns influenced by...
Length extrapolation algorithms based on Rotary position embedding (RoPE) have shown promising results in extending the context length of language models. However, understanding how can capture longer-range contextual information remains elusive. Based intuition that different dimensions correspond to frequency changes RoPE encoding, we conducted a dimension-level analysis investigate correlation between hidden dimension an attention head and its contribution capturing long-distance...
Long-term memory is significant for agents, in which insights play a crucial role. However, the emergence of irrelevant insight and lack general can greatly undermine effectiveness insight. To solve this problem, paper, we introduce Multi-Scale Insight Agent (MSI-Agent), an embodied agent designed to improve LLMs' planning decision-making ability by summarizing utilizing effectively across different scales. MSI achieves through experience selector, generator, selector. Leveraging three-part...
Long-Form Question Answering (LFQA) refers to generating in-depth, paragraph-level responses open-ended questions. Although lots of LFQA methods are developed, evaluating effectively and efficiently remains challenging due its high complexity cost. Therefore, there is no standard benchmark for evaluation till now. To address this gap, we make the first attempt by proposing a well-constructed, reference-based named Chinese exAmination Evaluation (CALF), aiming rigorously assess performance...
DarkSHINE is a newly proposed fixed-target experiment initiative to search for the invisible decay of Dark Photon via missing energy/momentum signatures, based on high repetition rate electron beam be deployed/delivered by Shanghai High XFEL and Extreme light facility (SHINE). This report elaborates baseline design introducing physics goals, experimental setups, details each sub-detector system technical designs, signal backgground modelings, expected sensitivities future prospects, which...
Model collapse in synthetic data indicates that iterative training on self-generated leads to a gradual decline performance. With the proliferation of AI models, will fundamentally reshape web ecosystem. Future GPT-$\{n\}$ models inevitably be trained blend and human-produced data. In this paper, we focus two questions: what is impact language model training, how synthesize without collapse? We first pre-train across different proportions data, revealing negative correlation between...
Abstract Turbulent dissipation rate ( ɛ ) is a crucial parameter in turbulence theory, and an essential component of higher‐order planetary boundary layer schemes for numerical weather prediction climate models. It most often modeled diagnostically based on the scaling ∝ e 3/2 / L , where are kinetic energy (TKE) size largest turbulent eddies, respectively. Utilizing three‐month‐long vertically‐extended observations accompanied by high resolution large‐eddy simulations, scaling‐based ‐models...