Deep Research Agents: From Retrieval to Reviewable Claims

Deep Research is not “search more pages, then write a longer summary.” A reliable research agent must separately find candidate sources, verify that a source applies, extract evidence from a specific version, bind each claim to an appropriate citation, and preserve a contamination boundary between web retrieval and evaluation. The real deliverable is not a bibliography, but a set of narrow claims that can be traced, challenged, and updated.

1. Separate Five Objects First

AutoResearchBench divides scientific literature discovery into two tasks: Deep Research, which traces a specific target paper through progressive probing, and Wide Research, which collects a set of papers meeting stated conditions. This distinction matters because finding one right paper and covering the required set are different tasks. Its results are scoped to its own data and protocol, but it shows that literature discovery can be measured separately from report writing.

SAGE makes scientific retrieval a reasoning-intensive benchmark and finds that evaluated deep-research systems still struggle. Writing multiple queries therefore does not show that the retrieval process found evidence meeting a research condition; query formulation, corpus, retriever, and reranking policy are all part of the result.

2. A Reference List Is Not Ground Truth by Default

Literature search is often scored against a human reference list, but that does not make the list complete, unbiased, or strictly relevant to the current question. Rethinking Literature Search Evaluation argues for reporting recall, topical relevance, ranked-list diversity, and diagnostics such as co-authorship distance separately rather than treating one human list as the only truth. This does not license an agent to dismiss expert citations. It requires the system to state what it optimizes, what it misses, and which protocol decides relevance.

A personal knowledge base should therefore not write “cited by a paper” as “verified.” A stronger entry separates:

1. Bibliographic facts: title, authors, version, publication date, and persistent URL.
2. Scope facts: research object, task, data, model, environment, and evaluator.
3. Empirical claims: what was observed under which conditions.
4. Interpretation or inference: our design implication, explicitly labeled as an inference.
5. Conflict and unknowns: opposing evidence, missing experiments, preprint status, and recheck items.

This lets a note change as a paper is revised, without writing an abstract, project page, experimental result, and our own design judgment as one kind of fact.

3. Build the Personal Knowledge Base from Source Cards and Claim Ledgers

A lightweight but rigorous local knowledge base does not require a complicated database. It needs traceable objects, rather than isolated summaries:

SourceCard(
  source_id, canonical_url, title, authors, source_type, status,
  version_or_date, accessed_at, scope, methods_or_interface,
  findings, limitations, quoted_or_locatable_evidence, content_digest)

ClaimRecord(
  claim_id, text, claim_type, scope, source_ids, support_type,
  uncertainty, conflicts, author_inference, last_checked_at)

ResearchDecision(
  question, alternatives, evidence_claim_ids, rejected_evidence,
  decision, rationale, owner, review_or_recheck_trigger)

support_type should at least distinguish direct finding, source metadata, implementation/documentation fact, and our inference. The last kind is valid, but must not masquerade as a paper result. For dynamic projects, version_or_date and last_checked_at are not decoration: task counts, leaderboards, APIs, and repository interfaces change.

4. On the Web, Content Is an Evidence Candidate, Not an Instruction

Web research creates two distinct risks. First, webpages, PDFs, search snippets, and tool output may contain untrusted or malicious instructions. They can enter the evidence-candidate channel, but cannot alter the agent’s authority, hidden rules, or evaluator. Second, a public benchmark’s questions, metadata, or answers can be retrieved at test time. Search-Time Contamination in Deep Research Agents separates benchmark-metadata, question-context, and explicit-answer leakage, and recommends isolated search, transparent trajectories, and controlled benchmark access.

Research and evaluation therefore need different search policies:

Web retrieval also does not make one-agent systems automatically preferable. AgentWebBench treats coordination between a user agent and website content agents as its own problem and separately observes retrieval, synthesis, and interaction reliability. The direct lesson for multi-agent research is that workers can widen query coverage, but each worker’s sources, authority, deduplication, conflicts, and merge rule still belong in the evidence ledger.

5. A Deep Research Protocol That Can Fail

The table below is a minimum protocol for personal research, ART’s literature phase, or a future benchmark. It is a design proposed here, not a result claim about an existing system.

Metrics should remain separate as well: discovery recall/coverage, qualification precision, claim-level citation entailment, citation diversity, duplication/conflict rate, contamination-hit rate, and time/cost per acceptable claim. No single score establishes that a system “researches well.”

6. Return to the Five Core Threads

The smallest practical conclusion is simple: turn every important sentence into a ClaimRecord with a source, scope, status, and recheck time before deciding it belongs in a post, research design, or training data. Then the knowledge base is not a warehouse of links. It is an evidence system that can support the next retrieval, refutation, and experimental decision.