Language models are now capable of solving tasks that require dealing with long sequences consisting of hundreds of thousands of tokens. However, they often fail on tasks that require repetitive use of simple rules, even on sequences that are much shorter than those seen during training. For example, state-of-the-art LLMs can find common items in two lists with up to 20 items but fail when lists have 80 items. In this paper, we introduce Lissard, a benchmark comprising seven tasks whose goal is to assess the ability of models to process and generate wide-range sequence lengths, requiring repetitive procedural execution. Our evaluation of open-source (Mistral-7B and Mixtral-8x7B) and proprietary models (GPT-3.5 and GPT-4) show a consistent decline in performance across all models as the complexity of the sequence increases. The datasets and code are available at https://github.com/unicamp-dl/Lissard

翻译：语言模型现能处理包含数十万词元的长序列任务，但面对需重复应用简单规则的任务时，即便序列长度远短于训练数据，仍常失败。例如，最先进的大语言模型能在两个最多20项的列表中找出共同项，但当列表包含80项时便无法完成。本文提出Lissard基准测试，包含七项任务，旨在评估模型处理与生成大范围序列长度、需重复程序执行的能力。我们对开源模型（Mistral-7B和Mixtral-8x7B）及商用模型（GPT-3.5和GPT-4）的评估表明，随着序列复杂度增加，所有模型性能均呈持续下降趋势。数据集与代码见https://github.com/unicamp-dl/Lissard