IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discovery
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908938643963904 |
|---|---|
| author | Sheth, Ivaxi Jin, Zhijing Wilder, Bryan Janzing, Dominik Fritz, Mario |
| author_facet | Sheth, Ivaxi Jin, Zhijing Wilder, Bryan Janzing, Dominik Fritz, Mario |
| contents | In the presence of confounding between an endogenous variable and the outcome, instrumental variables (IVs) are used to isolate the causal effect of the endogenous variable. Identifying valid instruments requires interdisciplinary knowledge, creativity, and contextual understanding, making it a non-trivial task. In this paper, we investigate whether large language models (LLMs) can aid in this task. We perform a two-stage evaluation framework. First, we test whether LLMs can recover well-established instruments from the literature, assessing their ability to replicate standard reasoning. Second, we evaluate whether LLMs can identify and avoid instruments that have been empirically or theoretically discredited. Building on these results, we introduce IV Co-Scientist, a multi-agent system that proposes, critiques, and refines IVs for a given treatment-outcome pair. We also introduce a statistical test to contextualize consistency in the absence of ground truth. Our results show the potential of LLMs to discover valid instrumental variables from a large observational database. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_07943 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discovery Sheth, Ivaxi Jin, Zhijing Wilder, Bryan Janzing, Dominik Fritz, Mario Artificial Intelligence In the presence of confounding between an endogenous variable and the outcome, instrumental variables (IVs) are used to isolate the causal effect of the endogenous variable. Identifying valid instruments requires interdisciplinary knowledge, creativity, and contextual understanding, making it a non-trivial task. In this paper, we investigate whether large language models (LLMs) can aid in this task. We perform a two-stage evaluation framework. First, we test whether LLMs can recover well-established instruments from the literature, assessing their ability to replicate standard reasoning. Second, we evaluate whether LLMs can identify and avoid instruments that have been empirically or theoretically discredited. Building on these results, we introduce IV Co-Scientist, a multi-agent system that proposes, critiques, and refines IVs for a given treatment-outcome pair. We also introduce a statistical test to contextualize consistency in the absence of ground truth. Our results show the potential of LLMs to discover valid instrumental variables from a large observational database. |
| title | IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discovery |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2602.07943 |