Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2605.00742 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911654274400256 |
|---|---|
| author | Papamarkou, Theodore Alquier, Pierre Bauer, Matthias Buntine, Wray Davison, Andrew Dziugaite, Gintare Karolina Filippone, Maurizio Foong, Andrew Y. K. Fortuin, Vincent Fouskakis, Dimitris Frellsen, Jes Hüllermeier, Eyke Karaletsos, Theofanis Khan, Mohammad Emtiyaz Kotelevskii, Nikita Lahlou, Salem Li, Yingzhen Liu, Fang Lyle, Clare Möllenhoff, Thomas Palla, Konstantina Panov, Maxim Sale, Yusuf Schweighofer, Kajetan Shelmanov, Artem Swaroop, Siddharth Trapp, Martin Waegeman, Willem Wilson, Andrew Gordon Zaytsev, Alexey |
| author_facet | Papamarkou, Theodore Alquier, Pierre Bauer, Matthias Buntine, Wray Davison, Andrew Dziugaite, Gintare Karolina Filippone, Maurizio Foong, Andrew Y. K. Fortuin, Vincent Fouskakis, Dimitris Frellsen, Jes Hüllermeier, Eyke Karaletsos, Theofanis Khan, Mohammad Emtiyaz Kotelevskii, Nikita Lahlou, Salem Li, Yingzhen Liu, Fang Lyle, Clare Möllenhoff, Thomas Palla, Konstantina Panov, Maxim Sale, Yusuf Schweighofer, Kajetan Shelmanov, Artem Swaroop, Siddharth Trapp, Martin Waegeman, Willem Wilson, Andrew Gordon Zaytsev, Alexey |
| contents | LLMs excel at predictive tasks and complex reasoning tasks, but many high-value deployments rely on decisions under uncertainty, for example, which tool to call, which expert to consult, or how many resources to invest. While the usefulness and feasibility of Bayesian approaches remain unclear for LLM inference, this position paper argues that the control layer of an agentic AI system (that orchestrates LLMs and tools) is a clear case where Bayesian principles should shine. Bayesian decision theory provides a framework for agentic systems that can help to maintain beliefs over task-relevant latent quantities, to update these beliefs from observed agentic and human-AI interactions, and to choose actions. Making LLMs themselves explicitly Bayesian belief-updating engines remains computationally intensive and conceptually nontrivial as a general modeling target. In contrast, this paper argues that coherent decision-making requires Bayesian principles at the orchestration level of the agentic system, not necessarily the LLM agent parameters. This paper articulates practical properties for Bayesian control that fit modern agentic AI systems and human-AI collaboration, and provides concrete examples and design patterns to illustrate how calibrated beliefs and utility-aware policies can improve agentic AI orchestration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_00742 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Position: agentic AI orchestration should be Bayes-consistent Papamarkou, Theodore Alquier, Pierre Bauer, Matthias Buntine, Wray Davison, Andrew Dziugaite, Gintare Karolina Filippone, Maurizio Foong, Andrew Y. K. Fortuin, Vincent Fouskakis, Dimitris Frellsen, Jes Hüllermeier, Eyke Karaletsos, Theofanis Khan, Mohammad Emtiyaz Kotelevskii, Nikita Lahlou, Salem Li, Yingzhen Liu, Fang Lyle, Clare Möllenhoff, Thomas Palla, Konstantina Panov, Maxim Sale, Yusuf Schweighofer, Kajetan Shelmanov, Artem Swaroop, Siddharth Trapp, Martin Waegeman, Willem Wilson, Andrew Gordon Zaytsev, Alexey Artificial Intelligence Machine Learning LLMs excel at predictive tasks and complex reasoning tasks, but many high-value deployments rely on decisions under uncertainty, for example, which tool to call, which expert to consult, or how many resources to invest. While the usefulness and feasibility of Bayesian approaches remain unclear for LLM inference, this position paper argues that the control layer of an agentic AI system (that orchestrates LLMs and tools) is a clear case where Bayesian principles should shine. Bayesian decision theory provides a framework for agentic systems that can help to maintain beliefs over task-relevant latent quantities, to update these beliefs from observed agentic and human-AI interactions, and to choose actions. Making LLMs themselves explicitly Bayesian belief-updating engines remains computationally intensive and conceptually nontrivial as a general modeling target. In contrast, this paper argues that coherent decision-making requires Bayesian principles at the orchestration level of the agentic system, not necessarily the LLM agent parameters. This paper articulates practical properties for Bayesian control that fit modern agentic AI systems and human-AI collaboration, and provides concrete examples and design patterns to illustrate how calibrated beliefs and utility-aware policies can improve agentic AI orchestration. |
| title | Position: agentic AI orchestration should be Bayes-consistent |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2605.00742 |