Spaces:
Running
Running
Commit History
-
Add E2E v3 + rawtext views
Verified signatureAdd the production overlay trajectory to run and Compare views, plus a structure alias for the cached E2E v3 scaffold. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: c2fe83fc-cc75-4d0b-b5ba-4a8def928fdcc34313e -
Polish cross-tab trajectory UX
Verified signaturePreserve qids across tabs, pair tool calls with results, compact gold answers, and collapse E2E supporting evidence. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: c2fe83fc-cc75-4d0b-b5ba-4a8def928fdc662f6d6 -
Add E2E structures, corrected c6, and corpus text
Verified signatureAdd E2E v3 constructed-structure shards, include corrected FinanceBench c6 in run and Compare tabs, and replace the CSP-blocked PDF iframe with lazy filing-text chunks. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: c2fe83fc-cc75-4d0b-b5ba-4a8def928fdccfbbab4
-
Add full-run baseline and E2E tabs
Verified signatureAdds c1, c2, naive-search, E2E v3, and Compare views over all 150 questions. The invalid retained c6 run remains excluded. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: c2fe83fc-cc75-4d0b-b5ba-4a8def928fdc9298a62
-
Add FinanceBench viewer: Corpus + Eval tabs
Verified signatureStatic viewer for the open-source FinanceBench dataset. - Corpus tab: 368 corpus PDFs, searchable dropdown by doc title, inline PDF render + metadata. - Eval tab: 150 annotated questions with gold answer, justification, reasoning, evidence spans, and doc link. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>f1e753d -
initial commit
Verified signaturef5f5021