[harness] nemo_retriever: Add dataset recall adapters for earnings and FinanceBench#1489
Merged
jioffe502 merged 4 commits intoNVIDIA:mainfrom Mar 5, 2026
Merged
Conversation
- normalize page-based CSVs into pdf_page recall keys - add financebench JSON adapter with pdf_only matching mode - recurse input globs so nested dataset directories ingest cleanly Signed-off-by: Jacob Ioffe <jioffe@nvidia.com> Made-with: Cursor
- reuse recall-core hit checks from batch pipeline - use adapter registry for cleaner recall adapter routing - parameterize recall-mode tests to reduce duplication Signed-off-by: Jacob Ioffe <jioffe@nvidia.com> Made-with: Cursor
- move nv_ingest_api import into NIM embed helper - avoid module import failure when API package is absent Signed-off-by: Jacob Ioffe <jioffe@nvidia.com> Made-with: Cursor
edknv
approved these changes
Mar 5, 2026
oliverholworthy
pushed a commit
to oliverholworthy/NeMo-Retriever
that referenced
this pull request
Mar 6, 2026
…d FinanceBench (NVIDIA#1489) Signed-off-by: Jacob Ioffe <jioffe@nvidia.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Adds dataset-specific recall input normalization and matching modes so FinanceBench and earnings can run end-to-end in the new retriever harness with correct recall semantics. Also simplifies internal recall logic by centralizing hit checks in recall core.
Adds recall_adapter + recall_match_mode wiring in harness config and run flow
Implements page_plus_one and financebench_json adapters for query normalization
Supports document-level (pdf_only) and page-level (pdf_page) recall with tests
Checklist