When Reasoning Meets Information Aggregation: A Case Study with Sports Narratives
This addresses the challenge of analytical reasoning in LLMs for sports data analysis, but it is incremental as it focuses on a specific domain and builds on existing evaluation methods.
The study tackled the problem of LLMs accurately aggregating information for reasoning by analyzing sports narratives, finding that most models, including GPT-4o, often fail to accurately aggregate basketball scores due to frequent scoring patterns, with open-source models like Llama-3 suffering from significant score hallucinations.
Reasoning is most powerful when an LLM accurately aggregates relevant information. We examine the critical role of information aggregation in reasoning by requiring the LLM to analyze sports narratives. To succeed at this task, an LLM must infer points from actions, identify related entities, attribute points accurately to players and teams, and compile key statistics to draw conclusions. We conduct comprehensive experiments with real NBA basketball data and present SportsGen, a new method to synthesize game narratives. By synthesizing data, we can rigorously evaluate LLMs' reasoning capabilities under complex scenarios with varying narrative lengths and density of information. Our findings show that most models, including GPT-4o, often fail to accurately aggregate basketball scores due to frequent scoring patterns. Open-source models like Llama-3 further suffer from significant score hallucinations. Finally, the effectiveness of reasoning is influenced by narrative complexity, information density, and domain-specific terms, highlighting the challenges in analytical reasoning tasks.