A5036645286- Phytochemical Studies and Bioactivities
- Imbalanced Data Classification Techniques
- Statistical and Computational Modeling
- Distributed and Parallel Computing Systems
- Robotics and Automated Systems
- Software Engineering Research
- Natural product bioactivities and synthesis
- Chromatography in Natural Products
Zhejiang Chinese Medical University
2020
We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as preliminary step, demonstrates remarkable capabilities. Through RL, naturally emerges with numerous powerful intriguing behaviors. However, it encounters challenges such poor readability, language mixing. To address these issues further enhance performance, we DeepSeek-R1, which incorporates...
We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, pioneers an auxiliary-loss-free strategy load balancing sets multi-token prediction training objective stronger performance. pre-train on 14.8 trillion diverse...
This study aimed to isolate, prepare and identify the main flavonoids from a standardized Smilax glabra extract (SGF) using preparative HPLC, MS, 1H NMR 13C NMR, determine contents of these UPLC, then compare their pharmacological activities in vitro. We obtained six SGF: astilbin (18.10%), neoastilbin (11.04%), isoastilbin (5.03%), neoisoastilbin (4.09%), engeletin (2.58%) (−)-epicatechin (1.77%). The antioxidant activity were evaluated by determining 2,2-diphenyl-1-picrylhydrazyl (DPPH)...