Run context
Pinned
Detect emerging, bridge-candidate, and undercited papers inside a curated audio-ML corpus, then expose the signals behind every recommendation.
Recommended
Momentum-weighted papers gaining relevance inside the current corpus.
Papers come from a materialized ranking run. The explanations reflect the recorded signal weights for the resolved run. The undercited family is scoped to a frozen low-citation candidate pool for the same corpus snapshot.
Run context
Pinned
Rows surfaced
20
Top score
0.595
Snapshot
Snapshot 2026-05-21
Showing 20 papers from bounded ML scorer order over materialized metadata; 20 include topic labels.
Topic labels are imported metadata and can be noisy; use them as coarse navigation hints, not authoritative classifications.
Focus paper: https://openalex.org/W4404396137 is not in the current top 20 rows for this view, but the run context is still pinned while you switch families.
Resolved run label: shadow-generalization-product-candidate-ranking-v1.
Ranking mode: bounded_ml_scorer. Order selected by bounded ML scorer; displayed scores/signals are materialized ranking metadata.
Bounded scorer serving is eligible only for Emerging requests with limit=20; other limits use materialized heuristic order.
Run label filter: shadow-generalization-product-candidate-ranking-v1. Resolved run: rank-83787b91ef; snapshot: source-snapshot-shadow-generalization-v1-20260521.
Storage provenance: materialized family rows are read from paper_scores. Result order was selected by the bounded ML scorer; displayed final_score and signal metadata still come from the materialized row. The undercited pool definition is documented in docs/candidate-pool-low-cite.md.
Live ranking surface
This response was ordered by the bounded ML scorer rollout. Displayed scores and signal explanations still come from the materialized ranking row.
Order selected by bounded ML scorer rollout. The bullets below describe materialized row metadata from the ranking run, not the final visible order.
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1667)
Recent attention: medium; used in final ranking (contribution to score: 0.1800)
Topic momentum: high; used in final ranking (contribution to score: 0.2480)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1639)
Recent attention: low; used in final ranking (contribution to score: 0.1000)
Topic momentum: high; used in final ranking (contribution to score: 0.2448)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1703)
Recent attention: low; used in final ranking (contribution to score: 0.0600)
Topic momentum: high; used in final ranking (contribution to score: 0.2715)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1646)
Recent attention: low; used in final ranking (contribution to score: 0.0800)
Topic momentum: high; used in final ranking (contribution to score: 0.2153)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1721)
Recent attention: low; used in final ranking (contribution to score: 0.0000)
Topic momentum: high; used in final ranking (contribution to score: 0.2591)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1714)
Recent attention: low; used in final ranking (contribution to score: 0.0000)
Topic momentum: high; used in final ranking (contribution to score: 0.2622)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1700)
Recent attention: low; used in final ranking (contribution to score: 0.0600)
Topic momentum: high; used in final ranking (contribution to score: 0.2448)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Reverse Engineering of Music Mixing Graphs With Differentiable Processors and Iterative Pruning
0.442Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1656)
Recent attention: low; used in final ranking (contribution to score: 0.0000)
Topic momentum: high; used in final ranking (contribution to score: 0.2765)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1689)
Recent attention: low; used in final ranking (contribution to score: 0.0900)
Topic momentum: high; used in final ranking (contribution to score: 0.2153)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1688)
Recent attention: high; used in final ranking (contribution to score: 0.3300)
Topic momentum: high; used in final ranking (contribution to score: 0.2294)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1764)
Recent attention: low; used in final ranking (contribution to score: 0.0600)
Topic momentum: high; used in final ranking (contribution to score: 0.2448)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1666)
Recent attention: low; used in final ranking (contribution to score: 0.0000)
Topic momentum: high; used in final ranking (contribution to score: 0.2465)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1664)
Recent attention: low; used in final ranking (contribution to score: 0.0900)
Topic momentum: high; used in final ranking (contribution to score: 0.2591)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1701)
Recent attention: low; used in final ranking (contribution to score: 0.0300)
Topic momentum: high; used in final ranking (contribution to score: 0.2153)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1689)
Recent attention: medium; used in final ranking (contribution to score: 0.1800)
Topic momentum: high; used in final ranking (contribution to score: 0.3000)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
High-Resolution Directivity Measurements of an Artificial Head and Mouth Shaped to Three Vowels
0.430Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1597)
Recent attention: low; used in final ranking (contribution to score: 0.0000)
Topic momentum: high; used in final ranking (contribution to score: 0.2700)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1646)
Recent attention: low; used in final ranking (contribution to score: 0.0000)
Topic momentum: high; used in final ranking (contribution to score: 0.2591)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1603)
Recent attention: low; used in final ranking (contribution to score: 0.0200)
Topic momentum: high; used in final ranking (contribution to score: 0.2448)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1530)
Recent attention: low; used in final ranking (contribution to score: 0.0900)
Topic momentum: high; used in final ranking (contribution to score: 0.2480)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging: embedding slice fit vs included-corpus centroid (title+abstract), plus citation velocity and topic growth; not universal relevance. Bridge signal not used here.
Embedding slice fit (corpus centroid): high; used in final ranking (contribution to score: 0.1608)
Recent attention: low; used in final ranking (contribution to score: 0.0300)
Topic momentum: high; used in final ranking (contribution to score: 0.2715)
Cross-cluster signal: not computed for this run
Similarity penalty: reduces score when non-zero (contribution to score: 0.0000)
Emerging can be reordered by the bounded ML scorer only when the backend rollout gate is enabled. Bridge and Under-cited are not served by that scorer; their order comes from materialized ranking rows.
The rule-only undercited list (/api/v1/recommendations/undercited) uses the same pool definition but is not tied to a corpus snapshot. For snapshot-scoped A/B against the ranked undercited family, use Evaluation.