# Scientific Hypothesis Generation

2 related articles

Subscribe to updates

researchsquare.com Oct 13, 2026
AI that reads the unpublished: a three-layer protocol flags four blind spots in CPEB biology and yields a testable sleep-forgetting hypothesis
AI systems for scientific discovery usually learn only from published positive results, leaving them structurally blind to failures and suppressed directions. A three-layer operational protocol turns that gap into executable steps: it scores theoretically possible but undocumented relationships in a knowledge graph, uses an adversarial classifier to surface features systematically suppressed in unpublished, rejected or failed data, and generates hypotheses that diverge from consensus yet pass viability filters. In a CPEB3/CPEB4 neurobiology case, it flagged four structural blind spots and produced a testable hypothesis about astrocytic translation during sleep and active forgetting.
AI systems for scientific discovery usually learn only from published positive results, leaving them structurally blind to failures and suppressed directions. A three-layer operational protocol turns that gap into executable steps: it scores theoretically possible but undocumented relationships in a knowledge graph, uses an adversarial classifier to surface features systematically suppressed in unpublished, rejected or failed data, and generates hypotheses that diverge from consensus yet pass viability filters. In a CPEB3/CPEB4 neurobiology case, it flagged four structural blind spots and produced a testable hypothesis about astrocytic translation during sleep and active forgetting.
Instituto Venezolano de Investigaciones Científicas
Lenin A. González-Paz（First author）
Negative Space Mining Dark Data Processing Scientific Hypothesis Generation Knowledge Graph Construction
arXiv Oct 6, 2026
In multi-agent hypothesis generation, the first self-critique round adds 34.5 pp of mechanism divergence while the equivalence rule itself shapes measured diversity
In a multi-agent hypothesis-generation workflow, the authors held opening hypotheses fixed and replayed downstream stages, using an LLM-as-a-judge to decide whether paired hypotheses describe the same mechanism: moving from 0 to 1 self-critique round produced 34.5 percentage points of extra mechanism-level divergence while 1 to 5 rounds added 1.3 pp; under controlled perturbations all three equivalence rules were invariant to meaning-preserving edits, but when the initiating event was replaced the LLM-as-a-judge read 83% of pairs as different mechanisms versus 0% for TF-IDF and 8% for embeddings, and changing only the rubric moved this rate from 38% to 96%.
In a multi-agent hypothesis-generation workflow, the authors held opening hypotheses fixed and replayed downstream stages, using an LLM-as-a-judge to decide whether paired hypotheses describe the same mechanism: moving from 0 to 1 self-critique round produced 34.5 percentage points of extra mechanism-level divergence while 1 to 5 rounds added 1.3 pp; under controlled perturbations all three equivalence rules were invariant to meaning-preserving edits, but when the initiating event was replaced the LLM-as-a-judge read 83% of pairs as different mechanisms versus 0% for TF-IDF and 8% for embeddings, and changing only the rubric moved this rate from 38% to 96%.
Johns Hopkins University Merck & Co., Inc., Rahway, NJ, USA (United States)
Ji Young Byun（First author）
Pairwise Equivalence Judgment LLM-as-a-Judge Self-Critique Scientific Hypothesis Generation

You’re all caught up

AI that reads the unpublished: a three-layer protocol flags four blind spots in CPEB biology and yields a testable sleep-forgetting hypothesis

In multi-agent hypothesis generation, the first self-critique round adds 34.5 pp of mechanism divergence while the equivalence rule itself shapes measured diversity