TRACE archives · Agentic Research Directory

TRACE archives

TRACE plans, searches, reviews and reranks historical archive sources. The author evaluation uses 1,752 questions over French documents from 1887; retrieval recall measures source discovery, not historical truth. Human checkpoints and expert-trust benefits remain future work. A public implementation is documented, but its match to the evaluated version and independent replication are unverified.

Workflow level n/a Source Peer-reviewed source Outcome Author-reported

Evidence & verification

Source level

Peer-reviewed source
Grounded in a peer-reviewed publication. This describes the source; it does not mean the reported results have been independently reproduced.

Outcome verification

Author-reported
Outcomes are reported by the team that built the system. The entry is source-checked against the team's own publications; not independently reproduced.

Sources checked

2026-10-19 (UTC)

Official CIKM 2026 programme corroborates acceptance of this exact paper (A67 Applied), with the event still upcoming. Technical claims were reviewed from arXiv2609.19897v1; final proceedings-version parity was not checked. Outcomes remain author-reported and do not establish historical interpretation quality or independent replication.

Source-checked does not mean independently reproduced. See the Method page for how these labels are assigned.

Sources (2)

Research applicability

Documented humanities research workflow, with source-version and evaluation limits retained.

Getting started

Read the primary manuscript and its limitations first. Check implementation access and terms separately; this review did not run an installation, API call to a research system, or experiment.

Maintenance

Source snapshot checked 2026-10-19. Current availability and experimental version parity are separate claims.

License

Research code/data/manuscript terms require separate verification; repository metadata alone is not a reuse clearance.

Workflow & human oversight

No workflow level assigned: the archival source-discovery algorithm. The researcher query/corpus is the input; expert checkpoints and trust/verification-effort benefits are proposed future work. Planning, retrieval, review and reranking are documented, but a completed human review/override pathway for the particular scholarly use is not established. Specific missing evidence: source-grounded operator validation/override and responsibility for final historical-source selection in an instantiated workflow. Review later primary workflow documentation before assigning a level; keep the bounded retrieval-method entry without an invented level.

Workflow level is not an evidence score; higher is not better evidence.

Approach

Corpus-directed decomposition, fused sparse/dense/date retrieval, bounded alternating search/review, accept/hold/reject, deferred reevaluation, briefing/replanning and ranked document merge. Expert intermediate checkpoints are future work.

Outcomes (as reported by the originating team)

1752questions over3386French1887documents/5664chunks:889single-hop,529generic,142bridge,192comparative. DefaultV4Flash R@10=.856,MRR=.653; V3.2 recall=.884. MRR averages reciprocal ranks of every gold document. Accepted precision=.393.

Limitations

Retrieval, not historical-interpretation quality. Baseline trajectory discovery-order ranking is asymmetric; single-year corpus/constants limit transfer. Internal24researcher/sixinstitution availability is author-reported, not user-study evidence; trust/verification-effort benefits remain untested.

Reproducibility

Author-reported; independent replication not established in this review. Documentary review only; no candidate code, model, simulation or experiment was executed. A public repository or result artifact is not proof of replication.

Adoption requirements

Untested suggestion: inspect the pinned sources and validate data, evaluator assumptions, implementation availability and researcher controls before planning a bounded trial. No setup or trial was executed.

Comments

No comments yet.

Add a comment

Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.