大语言模型中的时间感知：跨时间事实回忆基准测试
Time Awareness in Large Language Models: Benchmarking Fact Recall Across Time
AI总结：
本文构建覆盖2018–2024年8000多个日级标注事件的框架与数据集，并以TimeShift评估LLM时间事实回忆，发现基础模型常优于指令微调模型且大模型对改写事实仍脆弱。
摘要：
谁是美国总统？答案会随着提问时间而变化。尽管大语言模型（LLM）已在各类推理任务上接受评估，但它们往往遗漏了一个关键维度：时间。在现实场景中，答案的正确性常常与时间背景相关。为弥补这一空白，我们提出了一个新颖框架与数据集，涵盖2018年至2024年间的8000多个事件，具有日级粒度标注，并来源于全球政治、科学和商业等领域。我们的TimeShift评估方法系统地探测LLM的时间推理能力，结果显示，在时间敏感的事实回忆方面，基础模型往往优于指令微调和合成数据训练的对应模型。此外，我们发现即便是大规模模型，在处理改写后的事实时也表现出脆弱性，凸显了时间一致性方面尚未解决的挑战。通过识别这些局限，本工作为推动能够适应现实世界知识动态特性的时间感知语言模型迈出了重要一步。
英文摘要：
Who is the US President? The answer changes depending on when the question is asked. While large language models (LLMs) are evaluated on various reasoning tasks, they often miss a crucial dimension: time. In real-world scenarios, the correctness of answers is frequently tied to temporal context. To address this gap, we present a novel framework and dataset spanning over 8,000 events from 2018 to 2024, annotated with day-level granularity and sourced globally across domains such as politics, science, and business. Our TimeShift evaluation method systematically probes LLMs for temporal reasoning, revealing that base models often outperform instruction-tuned and synthetic-trained counterparts on time-sensitive recall. Additionally, we find that even large-scale models exhibit brittleness in handling paraphrased facts, highlighting unresolved challenges in temporal consistency. By identifying these limitations, our work provides a significant step toward advancing time-aware language models capable of adapting to the dynamic nature of real-world knowledge.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
