Finding gene function using LitMiner
Bibliographic record
Abstract
NRC (National Research Council, Canada) submitted 2 sets of results for the primary task in the TREC Genome track. The systems that generated these results were tuned primarily to achieve very high recall (above 90%) and secondarily to minimize the number of documents retrieved. Both submitted sets were the outputs of automatic systems (non-interactive, non-supervised) with a modular architecture. The TREC evaluation confirmed that recall for both submissions was extremely high: 543 out of 566 target documents (0.9594) were returned. In addition, these systems returned far fewer documents than were allowed by the genomic track rules. They returned an average of 196 documents per query across the 50 queries, with a median value of only 100 documents. For the first submission, the system was entirely based on Information Retrieval techniques, tuned to achieve very high recall and fair precision. Averaged precision was 0.3941 for the first submission. This first submission ranked third out of 49 runs submitted by all participants. For the second submission, reranking was done based on the outcome of an information extraction module, tuned towards the task of identifying gene function papers. This module identified 539 documents as highly promising; 121 of these turned out to be target documents, 418 weren't. All in all this caused the averaged precision to drop slightly to 0.3771- contrary to our expectations. This second submission ranked fifth out of all 49 runs. 1.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.005 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.006 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.008 | 0.006 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".