arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2501.17629cs.HCcs.AIcs.CY
收藏

一项严格的图灵测试:评估通用人工智能的基础

A Rigorous Turing Test: a Foundation for Evaluating Artificial General Intelligence

University of Canterbury(坎特伯雷大学);Oregon State University(俄勒冈州立大学)

Sharon Temtsin, Diane Proudfoot, David Kaber, Christoph Bartneck

AI总结:

该研究严格遵循图灵原始规则开展图灵测试,发现仅1名参与者误识别大型语言模型,表明其通过图灵测试的说法尚不成熟,图灵测试仍将是评估机器智能的核心手段。

摘要:

多项研究声称大型语言模型已通过图灵测试,因此能够“思考”,但 none 都未严格遵循图灵的原始指令。通过该测试具有重要意义,可作为机器展现类人智能的证据,以及商业和法律领域通用人工智能的标志。我们遵循图灵确定的指导原则,并在缺少详细说明的地方应用科学标准,使用大型语言模型开展了图灵的三人模仿游戏。我们进行了无时长限制的“计算机模仿人类”游戏,并将“男性模仿女性”游戏作为基准。在“计算机模仿人类”游戏中,仅有一名参与者误识别了该大型语言模型,这表明关于大型语言模型通过图灵测试的说法尚不成熟。两项任务的参与者均花费了超过五分钟时间,其中“男性模仿女性”游戏耗时更长;其他研究中更短的时间限制可能解释了早期得出的阳性结果。我们预计图灵测试在未来数年仍将是评估机器智能的核心手段。

英文摘要:

Several studies claim that large language models have passed the Turing Test and hence can "think", yet none follow Turing's original instructions precisely. Passing the test holds significance as evidence that a machine demonstrates human-like intelligence, and as a marker for artificial-general intelligence in commercial and legal domains. We conducted Turing's three-player imitation game with an LLM by following the guidelines identified by Turing and applying scientific standards wherever detailed instructions were missing. We performed a computer-imitates-human game without duration constraints and a man-imitates-woman game as a benchmark. In the computer-imitates-human game, only one participant misidentified the large language model, indicating that claims of large language models' passing the Turing test are premature. Participants required over five minutes for both tasks, with the man-imitates-woman game taking longer; shorter time limits elsewhere may explain earlier positive results. We expect the Turing Test to remain central in assessing machine intelligence for years to come.

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。

↑