通过视觉Grounding对抗语言漂移
Countering Language Drift via Visual Grounding
摘要:
紧急多智能体通信协议与自然语言有很大的不同,不容易被人类解释。我们发现,最初预先训练产生自然语言的代理也可能会经历有害的语言漂移:当在基于目标的任务中使用非语言奖励时,例如一些标量成功度量,通信协议可能很容易从根本上偏离自然语言。我们将翻译转换为多智能体通信游戏,并检查辅助训练约束在减轻语言漂移方面的有效性。我们表明,句法(语言模型可能性)和语义(视觉接地)的约束相结合,给出了最佳的通信性能,允许预先训练的代理保留英语语法,同时学习准确地传达预期的含义。
英文摘要:
Emergent multi-agent communication protocols are very different from natural language and not easily interpretable by humans. We find that agents that were initially pretrained to produce natural language can also experience detrimental language drift: when a non-linguistic reward is used in a goal-based task, e.g. some scalar success metric, the communication protocol may easily and radically diverge from natural language. We recast translation as a multi-agent communication game and examine auxiliary training constraints for their effectiveness in mitigating language drift. We show that a combination of syntactic (language model likelihood) and semantic (visual grounding) constraints gives the best communication performance, allowing pre-trained agents to retain English syntax while learning to accurately convey the intended meaning.
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。