Research map / Large language models / Instruction and fine-tuning
ParaTool: Shifting Tool Representations from Context to Parameters
Zekai Yu, Qi Meng, Qizhi Chu, Yu Hao, Chuan Shi, Cheng Yang
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 |
|---|---|---|---|
| encode each tool into its own LoRA parameters instead of providing docs and examples in context | Llama3.1-8B-Instruct · Stable ToolBench (average) | pass rate (%): 65.67 -> 69.57 | Improved |
| use per-tool parameters instead of a single global LoRA (ablation) | ParaTool · BFCL-V2 average | AST accuracy (%): 49.66 -> 76.23 (Live, Llama3.1-8B) | Improved |
| use soft tool composition instead of hard top-1 selection (ablation) | ParaTool · BFCL Live Parallel Multiple (Llama3.1-8B) | AST accuracy: drop of nearly 17% with Top-1 | Improved |
| use trained gating weights instead of averaging tool parameters (ablation) | ParaTool · BFCL (Llama3.1-8B) | AST accuracy | Improved |
| add parametric tool fine-tuning stage (ablation) | ParaTool · BFCL (Llama3.1-8B) | AST accuracy: 27.4% relative improvement | Improved |
| use oracle tool selector instead of learned gate (ablation) | ParaTool · BFCL Live Parallel Multiple | AST accuracy: gap of 6.0% | No clear effect |
| use learned gating network instead of retrieval-based tool selectors (ablation) | ParaTool · BFCL Parallel Multiple | AST accuracy: 31.7% relative gain over best selector | Improved |
| add more usage examples in the context | ICL tool calling · Llama3.1-8B and Qwen2.5-7B | tool calling accuracy | Mixed |
| parameterized tools instead of context | inference cost · Stable ToolBench | FLOPs: average reduction of 92.22% | Improved |
| fine-tune ToolLLaMA-style LoRA on all tools | Llama3.1-8B-Instruct · BFCL Non-live Parallel Multiple | AST accuracy (%): 85.50 -> 58.50 | Hurt |
Closest papers by meaning
- AutoTool: Efficient Tool Selection for Large Language Model Agents (AAAI 2026 · similarity 0.89)
- MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use (ICLR 2024 · similarity 0.89)
- Joint Optimization of Tool Creation and Use for Large Language Model Agents (NeurIPS 2026 · similarity 0.88)
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs (ICLR 2024 · similarity 0.88)
- ToolGen: Unified Tool Retrieval and Calling via Generation (ICLR 2025 · similarity 0.88)
- Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning (ICML 2026 · similarity 0.88)
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.
