Trajectory Supervision for Continual Tool-Use Learning in LLMs
Not on the map. Why?
Abstract (arXiv)
Most language-model training data shows final artifacts, not the process that produced them. We study a tractable version of this question in tool use: when a model learns a stream of new API domains, does keeping tool-use trajectories help compared with stripping the intermediate API trace? We fine-tune Llama 3.1 8B Instruct with QLoRA on API-Bank using four sequential domain blocks. Condition A strips previous API request/response lines from the prompt and trains the model to predict the next API call. Condition B keeps the trajectory context. In a single-seed pilot, full held-out generation evaluation shows that Condition B reaches 56.9% final exact full-call accuracy compared with 39.2% for Condition A. B also improves final API-name accuracy by 7.7 points. However, B uses 25.1% more training tokens, the run uses one seed, and the task is next-call prediction rather than full dialogue success.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 6 of its 6 references that are in the map. First posted on arXiv 10 May 2026.
Builds on
6 of 6 references.
References · 6
- 2024The Llama 3 Herd of Modelscited by 14,715
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
- 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
- 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,947
- 2023API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMscited by 591
- 2023TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Modelscited by 66
Built on by
Not cited yet.