评测大模型错误可预测性，助力更安全的AI部署。
PredictaBoard: Benchmarking LLM Score Predictability
- 构建协作式基准框架，评估预测器对模型错误的预判能力。
- 通过不同容忍误差下的拒绝率，量化模型预测可靠性。
- 推动开发更可预测的大模型，适合关注AI安全的研究者。
尽管大型语言模型具备出色能力，但在基础常识推理任务中仍表现出不可预测的失败，这种不确定性给安全部署带来重大挑战。为应对这一问题，我们提出PredictaBoard——一个新型协作式基准框架，用于评估评分预测器（称作评估者）能否从现有数据集的任务实例（即提示）中预判大模型的错误。该框架通过考察不同容忍误差下的拒绝率，评估大模型与评估者组合的表现。实验使用基线评估者和前沿大模型进行验证，结果表明需将可预测性与性能并重。PredictaBoard推动了更可靠评估体系的发展，使错误不仅被最小化，还能被提前识别和缓解。代码开源地址：https://github.com/Kinds-of-Intelligence-CFI/PredictaBoard。
原文摘要 · Abstract (English)
Despite possessing impressive skills, Large Language Models (LLMs) often fail unpredictably, demonstrating inconsistent success in even basic common sense reasoning tasks. This unpredictability poses a significant challenge to ensuring their safe deployment, as identifying and operating within a reliable "safe zone" is essential for mitigating risks. To address this, we present PredictaBoard, a novel collaborative benchmarking framework designed to evaluate the ability of score predictors (referred to as assessors) to anticipate LLM errors on specific task instances (i.e., prompts) from existing datasets. PredictaBoard evaluates pairs of LLMs and assessors by considering the rejection rate at different tolerance errors. As such, PredictaBoard stimulates research into developing better assessors and making LLMs more predictable, not only with a higher average performance. We conduct illustrative experiments using baseline assessors and state-of-the-art LLMs. PredictaBoard highlights the critical need to evaluate predictability alongside performance, paving the way for safer AI systems where errors are not only minimised but also anticipated and effectively mitigated. Code for our benchmark can be found at https://github.com/Kinds-of-Intelligence-CFI/PredictaBoard
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。