一篇新发表在arXiv上的研究探讨了“工程约束”（harness engineering）在提高大型语言模型在学术监督任务中可靠性方面的有效性。该研究将一个基准GPT-5聊天机器人与一个名为ASuS的系统进行了比较，ASuS使用了一个较小的GPT-4o-mini模型，但集成了检索、模式验证和LLM作为裁判循环等广泛的脚手架。结果表明，ASuS系统在多个维度上显著优于未经约束的GPT-5，这表明对于需要可靠性和可追溯性的任务，结构化工程比仅仅使用更大的模型更具影响力。 AI
影响 证明了结构化工程可以提高LLM的可靠性，挑战了特定应用中“模型越大越好”的范式。
排序理由 该集群包含一篇详细介绍LLM性能比较研究的论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2607.14707v1 Announce Type: cross Abstract: Large language models routinely produce fluent answers to single-shot prompts, yet deploying them as reliable components of a domain decision system is substantially harder. Closing this gap is the work of harness engineering: the…
Large language models routinely produce fluent answers to single-shot prompts, yet deploying them as reliable components of a domain decision system is substantially harder. Closing this gap is the work of harness engineering: the deliberate composition of deterministic scaffoldi…