AI RESEARCH
QUIET: A Multi-Blank Cascaded Story Cloze Benchmark for LLM Creative Generation Capability
arXiv CS.AI
•
ArXi:2605.25955v1 Announce Type: cross Large language models (LLMs) face a dual challenge in creative capability evaluation: existing benchmarks (e.g., Story Cloze Test, HellaSwag) measure models' discriminative ability over narrative continuation using multiple-choice recognition paradigms, rather than directly measuring creative generation capability; rubric-based scoring and LLM-as-Judge methods rely on subjective dimension assessment or natural language model outputs, and cannot provide objective, automated scoring mechanisms.