Don't Gamble, GAMBLe: An Analytical Framework for AI-Driven Research Systems

Computer Science > Artificial Intelligence [Submitted on 1 Jun 2026] Don't Gamble, GAMBLe: An Analytical Framework for AI-Driven Research Systems Marquita Ellis, Paul Castro AI-Driven Research Systems (ADRS) -- systems coupling LLMs with automated evaluation to discover algorithms, proofs, and designs -- are being optimized and adopted across domains, but the tools to analyze them have not kept pace. ADRS performance depends on component interactions that are poorly understood, expensive to explore, and (as we show) not well captured by standard convergence guarantees. These guarantees rely on structural assumptions that do not hold under the ADRS process we formalize. We introduce GAMBLe, a framework that decomposes ADRS behavior into four parameters (generator G, assessor \mathcal{A}, discovery mechanism \mathcal{M}, budget B) and one compositional object, the effective landscape L_{\text{eff}} = \mathcal{A} \circ G, which reveals that distinct generator-assessor pairs induce structurally different per-problem optimization landscapes. We exercise the framework on 760+ replicated runs (>46,000 iterations) spanning generators from single LLMs to dynamically-adaptive ensembles, mechanisms from greedy selection to co-evolutionary meta-search, and three NP-hard problems whose assessors range from continuous scoring to cliff functions. The experiments reveal no total ordering of generators or mechanisms: frontier models can underperform open-source alternatives and the simplest mechanism sometimes outperforms state-of-the-art meta-search. Results show that even under limited budgets (60 iterations per run), the right component choices can improve performance by 13-67% and search efficiency by 6-39x. Comments: Preprint. 21 pages (10 main, 11 appendix). 6 figures (2 in main, 4 in appendix) Subjects: Artificial Intelligence (cs.AI) ACM classes: I.2.8; I.2.6; I.2.4; I.2.11; G.1.6; F.2.2 Cite as: arXiv:2606.02863 [cs.AI]   (or arXiv:2606.02863v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2606.02863 Focus to learn more Submission history From: Marquita Ellis [view email] [v1] Mon, 1 Jun 2026 20:26:28 UTC (178 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-06 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)