A5003594074- Topic Modeling
- Natural Language Processing Techniques
- Access Control and Trust
- Spinal Hematomas and Complications
- Spinal Cord Injury Research
- Spinal Dysraphism and Malformations
- Multi-Agent Systems and Negotiation
- Subtitles and Audiovisual Media
- Adversarial Robustness in Machine Learning
- Advanced Malware Detection Techniques
China United Network Communications Group (China)
2024
Xiangyang Central Hospital
2022
Recently, the DeepSeek series of models, leveraging their exceptional reasoning capabilities and open-source strategy, is reshaping global AI landscape. Despite these advantages, they exhibit significant safety deficiencies. Research conducted by Robust Intelligence, a subsidiary Cisco, in collaboration with University Pennsylvania, revealed that DeepSeek-R1 has 100\% attack success rate when processing harmful prompts. Additionally, multiple companies research institutions have confirmed...
With the profound development of large language models(LLMs), their safety concerns have garnered increasing attention. However, there is a scarcity Chinese benchmarks for LLMs, and existing taxonomies are inadequate, lacking comprehensive detection capabilities in authentic scenarios. In this work, we introduce CHiSafetyBench, dedicated benchmark evaluating LLMs' identifying risky content refusing answering questions contexts. CHiSafetyBench incorporates dataset that covers hierarchical...
General large language models enhanced with supervised fine-tuning and reinforcement learning from human feedback are increasingly popular in academia industry as they generalize foundation to various practical tasks a prompt manner. To assist users selecting the best model application scenarios, i.e., choosing that meets requirements while minimizing cost, we introduce A-Eval, an application-driven LLMs evaluation benchmark for general models. First, categorize into five main categories 27...
The rapid growth of large language models(LLMs) has emerged as a prominent trend in the field artificial intelligence. However, current state-of-the-art LLMs are predominantly based on English. They encounter limitations when directly applied to tasks specific cultural domains, due deficiencies domain-specific knowledge and misunderstandings caused by differences values. To address this challenge, our paper proposes adaptation method for models contexts, which leverages instruction-tuning...
The spinal cord injury (SCI) is one of the major reasons causing motor dysfunctions patients. At present, few therapeutic strategies can effectively improve symptom SCI. Levetiracetam has been confirmed to alleviate nerve cells, while its functions in SCI remains unclear. In this study, C57BL/6J mice were used establish models observe effect levetiracetam on fed with 180 mg/kg when suffering from SCI, and Basso mouse score (BMS) CatWalk-assisted gait analysis mice. Nissl staining TUNEL...