aiqbench analysis
This research introduces CARE, an LLM-based controller for high-throughput chemistry that synthesizes executable scoring programs rather than issuing direct recommendations. By separating program synthesis from experiment selection and allowing logic updates without retraining model parameters, the method addresses critical issues of auditability and adaptability in sequential decision-making under budget constraints. The approach demonstrates superior performance on reaction optimization tasks by evolving its strategy based on new experimental outcomes while maintaining a reference policy for validation.
CARE: Context-Aware Ranking Evolution with Executable Scoring Programs for Budgeted Reaction Optimization
arXiv:2606.14581v4 Announce Type: replace-cross Abstract: High-throughput experimentation can evaluate many reaction conditions, yet combinatorial condition spaces still exceed the available experiment budget. This makes experiment selection a sequential decision problem: each new condition must be chosen from limited observations before its outcome is known. LLMs can express task-specific selection logic. A direct recommendation, however, is neither a persistent executable object that can be validated and revised nor an independently auditable decision rule. We introduce CARE, a reference-conditioned controller that separates program synthesis from experiment selection. An LLM writes an executable scoring program that ranks the remaining conditions, while a non-LLM reference policy supplies a numerical candidate and support summary. CARE forms an optional alternative from the program, applies a reference-conditioned intervention gate to compare it with the reference, and records the decision before the selected outcome is revealed. Each new outcome updates controller state and can trigger retention, revision, or regeneration of the active program. This outcome-guided program evolution changes the scoring logic without updating the LLM parameters. In matched offline replay with 30 seeds on eight reaction-optimization tasks, CARE attains the lowest normalized regret, the highest normalized best-so-far AUC, and the highest Top-1% Success@15 among the evaluated methods. These results support using a scoring program written by an LLM as one component of a reference-conditioned optimizer rather than as a standalone experiment selector.
Abstract:High-throughput experimentation can evaluate many reaction conditions, yet combinatorial condition spaces still exceed the available experiment budget. This makes experiment selection a sequential decision problem: each new condition must be chosen from limited observations before its outcome is known. LLMs can express task-specific selection logic. A direct recommendation, however, is neither a persistent executable object that can be validated and revised nor an independently auditable decision rule. We introduce CARE, a reference-conditioned controller that separates program synthesis from experiment selection. An LLM writes an executable scoring program that ranks the remaining conditions, while a non-LLM reference policy supplies a numerical candidate and support summary. CARE forms an optional alternative from the program, applies a reference-conditioned intervention gate to compare it with the reference, and records the decision before the selected outcome is revealed. Each new outcome updates controller state and can trigger retention, revision, or regeneration of the active program. This outcome-guided program evolution changes the scoring logic without updating the LLM parameters. In matched offline replay with 30 seeds on eight reaction-optimization tasks, CARE attains the lowest normalized regret, the highest normalized best-so-far AUC, and the highest Top-1% Success@15 among the evaluated methods. These results support using a scoring program written by an LLM as one component of a reference-conditioned optimizer rather than as a standalone experiment selector.
27 pages, 5 figures. Code: this https URL
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2606.14581 [cs.LG]
(or arXiv:2606.14581v4 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.14581 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 322 words
Related coverage
How Telcos Build Autonomous Networks with Agentic AI
NVIDIA · Jun 23, 2026
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · Aug 27, 2026
[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
Latent Space · Oct 13, 2026
Informe de riesgos de la IA de frontera (febrero–marzo de 2026)
METR · May 19, 2026