Research map / Large language models / Agents, code and math
AgentVocab: Structure-Aware Vocabulary Adaptation for Efficient LLM Agents
Kai Bian, Haosi Mo, Xuebo Liu, Shuangyong Song, Jing Li, Yongxiang Li, Min Zhang, Xuelong Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| adapt vocabulary with AgentVocab (800 structural + 200 content tokens) | Vanilla SFT Qwen2.5-7B-Instruct · tau-bench overall | latency (s): 0.209 -> 0.172 | Improved |
| adapt vocabulary with AgentVocab | Vanilla SFT Qwen2.5-7B-Instruct · tau2-bench overall | latency (s): 0.375 -> 0.302 | Improved |
| adapt vocabulary with AgentVocab | Vanilla SFT Qwen2.5-7B-Instruct · tau-bench overall | accuracy: 22.42% -> 24.65% | Improved |
| adapt vocabulary with AgentVocab | Vanilla SFT Qwen2.5-7B-Instruct · tau2-bench overall | accuracy: 23.79% -> 23.67% | No clear effect |
| adapt vocabulary with AgentVocab | Vanilla SFT Qwen2.5-7B-Instruct · tau2-bench Telecom | accuracy: 24.56% -> 17.25% | Hurt |
| add structural tokens only (800) (ablation) | Vanilla SFT · tau2-bench overall | latency (s): 0.375 -> 0.312 | Improved |
| add content tokens only (800) (ablation) | Vanilla SFT · tau2-bench overall | latency (s): 0.375 -> 0.360 | Mixed |
| increase structural budget from 800 to 1,000 (ablation) | AgentVocab structural-only · tau2-bench overall | accuracy: 23.42% -> 23.17% | No clear effect |
| reduce structural budget to 200 (ablation) | AgentVocab structural-only · tau2-bench overall | accuracy: 23.42% -> 21.00% | Hurt |
| add 200 content tokens to 800 structural (ablation) | AgentVocab · tau2-bench overall | accuracy: 23.42% -> 23.67% | Improved |
| increase content budget to 800 on top of 800 structural (ablation) | AgentVocab · tau2-bench overall | accuracy: 23.67% -> 20.20% | Hurt |
| fine-tune with SFT on tool traces | Qwen2.5-7B-Instruct · tau-bench overall | accuracy: 13.94% -> 22.42% | Improved |
Closest papers by meaning
- Offline Training of Language Model Agents with Functions as Learnable Weights (ICML 2024 · similarity 0.90)
- AgentSquare: Automatic LLM Agent Search in Modular Design Space (ICLR 2025 · similarity 0.90)
- ToolACE: Winning the Points of LLM Function Calling (ICLR 2025 · similarity 0.90)
- AutoTool: Efficient Tool Selection for Large Language Model Agents (AAAI 2026 · similarity 0.90)
- AgentBench: Evaluating LLMs as Agents (ICLR 2024 · similarity 0.89)
- Test-Time Adaptation for LLM Agents via Environment Interaction (ICLR 2026 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
