Large Language Models (LLMs) have excelled in multi-hop question-answering (M-QA) due to their advanced reasoning abilities. However, the impact of the inherent reasoning structures on LLM M-QA performance remains unclear, largely due to the absence of QA datasets that provide fine-grained reasoning structures. To address this gap, we introduce the Graph Reasoning-Structured Question Answering Dataset (GRS-QA), which includes both semantic contexts and reasoning structures for QA pairs. Unlike existing M-QA datasets, where different reasoning structures are entangled together, GRS-QA explicitly captures intricate reasoning pathways by constructing reasoning graphs, where nodes represent textual contexts and edges denote logical flows. These reasoning graphs of different structures enable a fine-grained evaluation of LLM reasoning capabilities across various reasoning structures. Our empirical analysis reveals that LLMs perform differently when handling questions with varying reasoning structures. This finding facilitates the exploration of textual structures as compared with semantics.

Old-school purists might argue that the pervasiveness of online chess has ruined the game's tactile appeal. Instead of pushing beautifully carved pieces across a painted board, most players nowadays drag and drop computer icons--a mouseclick is the only tangible element. But like something out of "Harry Potter," Square Off uses motors and magnets to glide an opponent's solid pieces over its rosewood surface, restoring an experience many players thought might be lost to technology. I bought two boards and shipped one to Danny. Paired to our smartphones via Bluetooth, the boards speak to one another over the internet.