聚合分数掩盖了模型升级中的细节退化，逐项回归分析才是迁移评估的关键。
arXiv:2608.17719v1 Announce Type: cross Abstract: Context: Software systems that depend on commercial large language model APIs must migrate to succes…
用公平赌局的平均结果重新定义评估指标，揭示评测背后的数学本质，适合关注模型评估理论的研究者。
arXiv:2401.14483v4 Announce Type: replace Abstract: In the current practices of machine learning, the evaluation of forecasts has become a cornerstone…
14种自动摘要指标与LLM评估器的大规模元评估，揭示跨领域长文档评估挑战，为反思式总结提供新框架。
arXiv:2604.25665v2 Announce Type: replace-cross Abstract: Reliable evaluation of large language model (LLM)-generated summaries remains an open challe…
一篇剖析AI暴露评分指标的学术论文，揭示其度量盲区与未来改进方向，适合关注AI影响评估的研究者与从业者。
arXiv:2606.23633v1 Announce Type: new Abstract: A set of exposure scores calculated in 2023 has become a central empirical input to the future of work…
量化模型部署的保真度评估，揭示位移≠方向的关键误区，助你选对指标。
arXiv:2606.19558v1 Announce Type: new Abstract: Fidelity metrics, such as per-token KL divergence (KLD) against a high-precision reference, are often …
挑战无监督单词发现中传统评价指标，提出新视角下的词典评估方法。
arXiv:2606.06183v1 Announce Type: cross Abstract: Building a lexicon from discovered word-like units is a central goal in zero-resource speech process…
提出冗余感知RLVR方法，改进多样本代码生成的评估指标
arXiv:2605.28022v1 Announce Type: new Abstract: LLMs for code generation are commonly evaluated in repeated-sampling settings using Pass@k, where mult…
新指标WCS精准测量LLM词汇可达性，揭示采样策略如何限制输出词汇多样性，为模型评估提供新视角。
arXiv:2605.27268v1 Announce Type: cross Abstract: Modern Large Language Models (LLMs) are often criticized for producing repetitive and homogeneous te…
用分布微调技术让LLM写作告别公式化，创造力提升164%，效果显著
Article URL: https://rosmine.ai/2026/05/18/fixing-llm-writing-with-distribution-fine-tuning/ Comments URL: https://news.ycombinator.com/item?id=481847…