QUIET: A Multi-Blank Cascaded Story Cloze Benchmark for LLM Creative Generation Capability 事件

QUIET: A Multi-Blank Cascaded Story Cloze Benchmark for LLM Creative Generation Capability arXiv:2605.25955v1 Announce Type: new Abstract: 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 metho