Controlling Tool Use with Heading-Specific Activation Steering
Quick Answer
This study reveals that steering vectors from heading-anchors can effectively control tool invocation in large language models, reducing unnecessary tool use across five open-source models.
Quick Take
This study reveals that steering vectors from heading-anchors can effectively control tool invocation in large language models, reducing unnecessary across five open-source models. The research highlights a non-linear relationship between tool-invocation behavior and the suppression vector, suggesting distinct internal signatures for different tool types.
Key Points
- Steering vectors exert bidirectional control over tool-invocation behavior.
- Effective suppression of unnecessary tool use occurs where parametric reasoning suffices.
- Tool-invocation steps show diffuse, bimodal alignment with suppression vectors.
- Distinct internal signatures exist for different tool types with low feature overlap.
- Geometric properties indicate the non-parametric nature of tools.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Tool-augmented large language models extend their capabilities beyond parametric knowledge through external tools, but tend to invoke them unnecessarily. We investigate whether tool-use decisions have any stable internal representation that can be extracted and manipulated, a question that is non-trivial given that tools exist entirely in context at inference time and have no direct encoding in model weights. We show that steering vectors extracted from heading-anchors positions exert bidirectional causal control over tool-invocation behavior across five open-source models and three domains, suppressing unnecessary tool use most effectively in domains where parametric reasoning suffices. However, geometric analysis reveals that this causal effectiveness does not correspond to clean linear structure: tool-invocation steps exhibit diffuse, bimodal alignment with the suppression vector rather than the consistent negative alignment a linear encoding account would predict, and different tool types recruit largely distinct internal signatures with low cross-tool feature overlap. We hypothesize these geometric properties are indicative of the non-parametric nature of tools, and distinguish tool-use steering vectors from those extracted for parametrically grounded concepts. The relationship between this geometric irregularity and the observed causal effectiveness remains an open question.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.05790 [cs.AI] |
| (or arXiv:2607.05790v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.05790 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yuqi Chen [view email]
[v1]
Tue, 7 Jul 2026 03:34:42 UTC (1,248 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →Onnes: A Physics-Grounded LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure
Onnes is a physics-grounded digital twin simulator for dilution refrigerators, enhancing cryogenic fault diagnosis in quantum computing. It achieves a classification accuracy of 99.0% using few-shot demonstrations, matching a supervised ML classifier, while maintaining a low false alarm rate of 6.4% on real hardware.