Research map / Large language models / Retrieval and knowledge
Interactive Rare-Category-of-Interest Mining from Large Datasets
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use IRim RCD (log-time queries) instead of Interleave | IRim · Kddcup RCD query | response time (s): 76.960 -> 0.762 | Helped |
| use IRim RCD instead of NNDM | IRim · Shuttle RCD query | response time (s): 393.894 -> 0.133 | Helped |
| use IRim RCD instead of HMS | IRim · Shuttle RCD query | response time (s): 5406.318 -> 0.133 | Helped |
| use IRim RCD instead of Clover | IRim · Shuttle RCD query | response time (s): 1684.139 -> 0.133 | Helped |
| use IRim RCD instead of Interleave | IRim · Game RCD query | response time (s): 410.663 -> 0.568 | Helped |
| use IRim RCE instead of FRANK | IRim · Game RCE ranking round | response time (s): 165.9 -> 0.577 | Helped |
| use IRim RCE instead of NN model | IRim · Kddcup RCE ranking round | response time (s): 3.769 -> 0.498 | Helped |
| use IRim RCE instead of RACH | IRim · Shuttle RCE ranking round | response time (s): 91.549 -> 0.284 | Helped |
| collaborative reconstruction using positive and negative neighbors vs best existing RCE method | IRim · ten rare categories on five datasets | accuracy: 11.75% improvement | Helped |
| learn from negative feedback in addition to compactness assumption (vs CC, NN, FRANK) | IRim RCE · non-separable example scenario | accuracy | Helped |
| stable areas in parameter setting panorama so only one setting per area is tried | IRim RCD · five datasets, 17 participants | number of queries to detect a seed | Helped |
| dataset size grows for baselines with O(n^2) query cost | existing RCD methods · Abalone to Kddcup | response time | Hurt |
Closest papers by meaning
- Prior-Constrained Association Learning for Fine-Grained Generalized Category Discovery (AAAI 2025 · similarity 0.80)
- Prototypical Hash Encoding for On-the-Fly Fine-Grained Category Discovery (NeurIPS 2024 · similarity 0.80)
- Happy: A Debiased Learning Framework for Continual Generalized Category Discovery (NeurIPS 2024 · similarity 0.80)
- Active Generalized Category Discovery (CVPR 2024 · similarity 0.79)
- Generalized Category Discovery under Domain Shift: A Frequency Domain Perspective (NeurIPS 2025 · similarity 0.79)
- A Unified Knowledge Transfer Network for Generalized Category Discovery (AAAI 2024 · similarity 0.78)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.