Tag: LLM
All the papers with the tag "LLM".
Procedural Memory Is Not All You Need: Bridging Cognitive Gaps in LLM-Based Agents
grok-3-latestScore: 0.77Published: at 11:18本文提出一种模块化架构,通过分离程序性、语义和关联功能,弥补大型语言模型在复杂‘恶劣’学习环境中的认知缺陷,为构建适应性AI代理提供了理论基础。
Procedural Memory Is Not All You Need: Bridging Cognitive Gaps in LLM-Based Agents
grok-3-latestScore: 0.77Published: at 11:18本文提出一种模块化架构,通过解耦程序性、语义和关联功能,弥补大型语言模型在‘恶劣’学习环境中的认知缺陷,为构建适应复杂现实世界的 AI 代理提供了理论框架。
Lightweight Clinical Decision Support System using QLoRA-Fine-Tuned LLMs and Retrieval-Augmented Generation
grok-3-latestScore: 0.52Published: at 10:31本文提出了一种轻量化的临床决策支持系统,通过结合检索增强生成(RAG)和量化低秩适应(QLoRA)微调技术,提升了大型语言模型在医疗任务中的准确性和效率,同时降低了计算资源需求。
Automatic Calibration for Membership Inference Attack on Large Language Models
grok-3-latestScore: 0.58Published: at 10:15本文提出自动校准成员推断攻击(ACMIA)框架,通过温度调整校准大型语言模型输出概率,显著提高成员推断精度并降低假阳性率,无需外部参考模型。
Validating the Effectiveness of a Large Language Model-based Approach for Identifying Children's Development across Various Free Play Settings in Kindergarten
grok-3-latestScore: 0.50Published: at 09:40本文提出并验证了一种基于大型语言模型和学习分析的方法,通过分析幼儿园儿童自由玩耍自述,准确评估其多维度发展能力,为教育者提供数据驱动的个性化洞察。
Avoid Recommending Out-of-Domain Items: Constrained Generative Recommendation with LLMs
grok-3-latestScore: 0.55Published: at 09:08本文提出 RecLM-cgen 方法,通过约束生成技术消除大型语言模型在推荐系统中的领域外物品推荐问题,同时显著提升推荐准确性,为生成式推荐提供轻量级实用解决方案。