GISTBench: Evaluating LLM User Understanding via Evidence-Based Interest Verification
This work addresses the need for better evaluation of LLM user understanding in recommendation systems, though it is incremental as it builds on existing benchmarks by focusing on interest verification rather than item prediction.
The authors tackled the problem of evaluating LLMs' ability to understand users from interaction histories in recommendation systems, introducing GISTBench with novel metrics and a synthetic dataset, and found performance bottlenecks in current LLMs, such as limited accuracy in counting and attributing engagement signals.
We introduce GISTBench, a benchmark for evaluating Large Language Models' (LLMs) ability to understand users from their interaction histories in recommendation systems. Unlike traditional RecSys benchmarks that focus on item prediction accuracy, our benchmark evaluates how well LLMs can extract and verify user interests from engagement data. We propose two novel metric families: Interest Groundedness (IG), decomposed into precision and recall components to separately penalize hallucinated interest categories and reward coverage, and Interest Specificity (IS), which assesses the distinctiveness of verified LLM-predicted user profiles. We release a synthetic dataset constructed on real user interactions on a global short-form video platform. Our dataset contains both implicit and explicit engagement signals and rich textual descriptions. We validate our dataset fidelity against user surveys, and evaluate eight open-weight LLMs spanning 7B to 120B parameters. Our findings reveal performance bottlenecks in current LLMs, particularly their limited ability to accurately count and attribute engagement signals across heterogeneous interaction types.