English(EN)Translate-R1: Cost-Aware Translation Tool Use via Reinforcement Learning
LLM智能体通过技能重写和翻译策略优化成本
作者PulseAugur 编辑部·[6 个来源]·
研究人员正在探索大型语言模型智能体的成本感知策略，以提高效率和性能。一篇论文介绍了一个技能重写框架，该框架通过保留关键操作锚点来优化成本，从而降低了智能体成本。另一项研究侧重于成本感知的翻译工具使用，开发了一种强化学习策略，该策略能够智能地决定何时翻译输入，以利用LLM的能力而不产生不必要的费用，特别有利于低资源语言。第三篇论文提出了一个用于机器翻译源重写的强化学习框架，该框架直接优化下游翻译质量，性能优于基于提示的方法。
AI
arXiv:2606.09421v2 Announce Type: replace Abstract: Large language model agents increasingly rely on skills: reusable procedural documents encoding workflows, tool use, implementation patterns, validation checks, and domain rules. Skill rewriting is often treated as prompt compre…
arXiv:2606.08011v1 Announce Type: cross Abstract: Although directly prompting off-the-shelf Large Language Models (LLMs) to generate meaning-preserving source rewrites can effectively enhance Machine Translation (MT) quality, doing so requires manually tuning prompts for differen…
Large language model agents increasingly rely on skills: reusable procedural documents encoding workflows, tool use, implementation patterns, validation checks, and domain rules. Skill rewriting is often treated as prompt compression, but shorter skills can make agents more expen…
arXiv:2606.06835v1 Announce Type: new Abstract: The performance gap across languages in LLMs is well documented, and closing it natively requires pretraining or fine-tuning on corpora that, for most languages, do not exist. Translation offers an alternative: converting an input i…
Although directly prompting off-the-shelf Large Language Models (LLMs) to generate meaning-preserving source rewrites can effectively enhance Machine Translation (MT) quality, doing so requires manually tuning prompts for different MT models. In this work, we propose RLSR (Reinfo…
The performance gap across languages in LLMs is well documented, and closing it natively requires pretraining or fine-tuning on corpora that, for most languages, do not exist. Translation offers an alternative: converting an input into the model's dominant language unlocks its fu…