Search papers, labs, and topics across Lattice.
The paper introduces Planner-Caller-Generator (P-C-G), a small-scale language model (SLM) based agent architecture designed for improved Korean tool use by separating planning, calling, and generation roles. P-C-G incorporates a Korean-first value policy to mitigate code-switching related execution failures common in Korean settings. Experiments evaluating P-C-G across various tool-use scenarios demonstrate competitive accuracy and end-to-end quality while reducing token usage and maintaining acceptable latency, suggesting the effectiveness of role-specialized SLMs for this task.
Forget LLMs, role-specialized small language models can nail Korean tool use with competitive accuracy and lower costs.
We propose a small-scale language model (SLM) based agent architecture, Planner-Caller-Generator (P-C-G), optimized for Korean tool use. P-C-G separates planning, calling, and generation by role: the Planner produces an initial batch plan with limited on-demand replanning; the Caller returns a normalized call object after joint schema-value validation; and the Generator integrates tool outputs to produce the final answer. We apply a Korean-first value policy to reduce execution failures caused by frequent Korean-to-English code switching in Korean settings. Evaluation assumes Korean queries and Korean tool/parameter specifications; it covers single-chain, multi-chain, missing-parameters, and missing-functions scenarios, and is conducted via an LLM-as-a-Judge protocol averaged over five runs under a unified I/O interface. Results show that P-C-G delivers competitive tool-use accuracy and end-to-end quality while reducing tokens and maintaining acceptable latency, indicating that role-specialized SLMs are a cost-effective alternative for Korean tool-use agents.
