Tag: Large Language Model
All the articles with the tag "Large Language Model".
-
Communicating Activations Between Language Model Agents
This paper introduces Activation Communication (AC), a novel method for inter-LLM communication using intermediate activations instead of natural language, achieving up to 27% performance improvement over traditional methods with significantly reduced compute across coordination games and reasoning benchmarks.
-
Merge to Mix: Mixing Datasets via Model Merging
本文提出*Merge to Mix*方法,通过模型合并技术作为代理,高效选择数据集混合用于大型模型微调,在图像分类和语言任务中显著优于传统方法,接近甚至部分超过Oracle性能。
-
MELoRA: Mini-Ensemble Low-Rank Adapters for Parameter-Efficient Fine-Tuning
本文提出MELoRA,通过并行堆叠多个小型LoRA模块实现更高的等效秩,以更少的参数在自然语言理解和指令跟随任务上显著优于LoRA。
-
How much do language models memorize?
本文提出了一种基于信息论的记忆量化方法,通过区分无意记忆和泛化,测量GPT风格语言模型的容量约为每个参数3.6比特,并揭示了数据集规模与模型容量比对双重下降和成员推断性能的影响。
-
Enhancing Efficiency and Exploration in Reinforcement Learning for LLMs
本文提出动态采样预算分配和温度调度机制,通过基于问题难度的资源再分配和维持策略熵的探索能力,显著提升了大型语言模型在数学任务中的强化学习效率和性能,尤其在AIME 2024基准上pass@1和pass@16分别提高5.31%和3.33%。