In the realm of scientific discovery, LLM systems are increasingly utilized for ideation, literature synthesis, experiment planning, and report generation. However, the first research questions they propose often remain difficult to audit, as these questions may sound plausible without revealing the mechanisms, falsifiers, or assumptions that a scientist should inspect. To address this, we introduce FirstResearch, a first-principles research question formation framework, whose core artifact is a structured Research Question Certificate. This certificate records primitive definitions, assumptions, a mechanism model, a tension or contradiction, a falsifiable hypothesis, a minimal decisive test, and a failure update rule, making the proposed question inspectable before downstream execution.
On ten LLM-agent research topics, FirstResearch outperforms controlled prompt-level baselines inspired by AI co-scientist, Agent Laboratory, and AI Scientist-v2 under a primary DeepSeek-blind-judge protocol. A Gemini-2.5-Flash independent-judge rescore of the same 40 baseline packages preserves the system-level ranking, with FirstResearch scoring 4.86/5 versus 4.38/5 for the strongest baseline and an average Pearson agreement of 0.865. A one-repeat ablation checkpoint further suggests that the certificate-centered core is the strongest component: certificate-only scoring reaches 4.90/5 under DeepSeek and 4.88/5 under Gemini, while removing certificates drops below 1/5 under both judges. These results are preliminary and use LLM judges rather than human domain experts, but they support a narrow scientific-discovery claim: explicit derivation constraints are a promising mechanism for making LLM-generated scientific questions more auditable.
Code, prompts, saved outputs, and reproduction scripts are available at GitHub.
Blogger's Review: FirstResearch provides robust support for the auditability of research questions, with its structured certificate design enhancing transparency and offering greater credibility to the scientific discovery process. This framework is poised to lead new directions in the application of LLMs in scientific research.