Optimist Deep Research: An Iterative, State-Centric Agent for Autonomous Literature Research
Optimist Deep Research is an open-source iterative agent that audits literature search through a mutable evidence state.
Optimist Deep Research is an open-source literature-research agent that iterates over an explicit, mutable research state rather than a single retrieve-then-summarize step. It decomposes questions into subquestions, queries arXiv, OpenAlex, Crossref, PubMed, Semantic Scholar, and the web, then deduplicates results and expands them by citation snowballing. Documents are read into a structured evidence store, a critic can revise the research plan, and the report cites only retrieved documents. A case study read 28 of 237 retrieved documents in about three and a half minutes and also exposed weaknesses.
- Agent maintains a mutable tree of research subquestions.
- Queries cover arXiv, OpenAlex, Crossref, PubMed, Semantic Scholar, and the web.
- Four-tier deduplication and citation snowballing build a provenance registry.
- Case study read 28 of 237 documents in about 3.5 minutes.
- Organizations
- arXivOpenAlexCrossrefPubMedSemantic Scholar
- AI models
- Optimist Deep Research
From the original article
1 min read at arxiv.orgMost retrieval-augmented systems treat literature search as a single retrieval step followed by summarization. Real literature research works differently: the investigator's understanding of the question changes as they read, new subquestions appear, and evidence is weighed, not just collected.
By Gihan Panapitiya · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.