Use when the user wants to synthesize findings across multiple papers into a coherent narrative, comparison, or temporal view.
Synthesizes findings across multiple papers into a coherent narrative, structured comparison, or integrated model. Goes beyond summarizing individual papers to produce genuinely novel synthesis — identifying patterns, progressions, and insights that only emerge when reading across the corpus as a whole.
synthesis = cross_paper_synthesis.narrative(
papers=review.get_papers(tag="attention-mechanisms"),
focus="methodological evolution",
max_words=800,
audience="expert"
)
cross_paper_synthesis.comparison_table(
papers=review.get_papers(tag="protein-lm-benchmarks"),
dimensions=["model_size", "training_data", "benchmark", "metric", "score"],
format="markdown_table"
)
cross_paper_synthesis.timeline(
topic="in-context learning capabilities",
papers=review.get_papers(),
group_by="year"
)
Narrative synthesis is prose with inline citations. Comparison tables are Markdown or LaTeX. Timeline output shows year-by-year progression with key papers anchoring each phase.
related_work_scout for related-work drafting and literature synthesispaper_editor when rewriting a weak compare-and-contrast section