Weekly Papers

Mixture-of-Agents Enhances Large Language Model Capabilities Mixture-of-Agents 增强大语言模型能力 最近在大语言模型 (L

Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Du

Meteor: Mamba-based Traversal of Rationale for Large Language and Vision Models Meteor: 基于 Mamba 的大型

PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning PLLaVA：图像至视

AutoCrawler: A Progressive Understanding Web Agent for Web Crawler Generation AutoCrawler: 一个为 Web 爬

Pre-training Small Base LMs with Fewer Tokens 使用更少的 Token 对小型基础 LMs 进行预训练 我们研究了一种简单方法来开发一个小型基础语言模型 (

Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences 直接纳什优化：教

Jamba: A Hybrid Transformer-Mamba Language Model Jamba：混合 Transformer-Mamba 语言模型 我们推出了 Jamba，这是一种基于创

Can large language models explore in-context? 大语言模型能进行上下文探索吗？ 我们研究了当代大语言模型（LLMs）在多大程度上能够进行探索，这在强化学习和

Uni-SMART: Universal Science Multimodal Analysis and Research Transformer Uni-SMART: 通用科学的多模态分析研究 Tr

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context Gemini 1.5：跨数百万T

VisionLLaMA: A Unified LLaMA Interface for Vision Tasks VisionLLaMA: 视觉任务的统一 LLaMA 接口 大语言模型基于 Transf

MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases MobileLLM: 针对移动设

Linear Transformers with Learnable Kernel Functions are Better In-Context Models 具有可学习核函数的线性 Transfo

Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning Aya 数据集：多语言指令微调的公开获取资源 数据