Research map / Learning theory and algorithms / Prediction and decision losses
Group Testing for Accurate and Efficient Range-Based Near Neighbor Search for Plagiarism Detection
Harsh Shah, Kashish Mittal, Ajit Rajwade
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 binary-splitting group testing instead of exhaustive search for range NN search | Our-Sum · MIRFLICKR, rho=0.8 | mean query time (ms): 1222 -> 54.0 | Improved |
| use binary-splitting group testing instead of exhaustive search | Our-Sum · IMDB-Wiki, rho=0.8 | mean query time (ms): 800 -> 98.8 | Improved |
| sum pooling instead of element-wise max pooling (ablation) | binary splitting · MIRFLICKR, rho=0.8 | mean query time (ms): 81 -> 54.0 | Improved |
| sum pooling instead of element-wise max pooling (ablation) | binary splitting · ImageNet, rho=0.8 | dot products per query: 12339 (sum) vs 5713 (max) | Hurt |
| use Our-Sum instead of Flinng / Flann-K approximate group testing | range NN search · MIRFLICKR, rho=0.8 | mean recall: 0.268 -> 1.0 (Flinng) | Improved |
| use Our-Sum instead of Hnsw in streaming insertion | streaming setting · ImageNet, streaming | mean insertion time (ms): 236 -> 0.28 | Improved |
| use Scann instead of Our-Sum | range NN search · MIRFLICKR, rho=0.8 | mean query time (ms): 54.0 -> 23 (recall 1.0 -> 0.996) | Improved |
| use Hnsw instead of Our-Sum on IMDB-Wiki | range NN search · IMDB-Wiki, rho=0.8 | mean query time (ms): 98.8 -> 1076 | Hurt |
| apply compressed-sensing expander recovery to approximate NN search | group testing NN search · simulations | recovery error bound | Hurt |
Closest papers by meaning
- Practical Near Neighbor Search via Group Testing (NeurIPS 2021 · similarity 0.88)
- Searching for and Avoiding Hidden Sets Using Queries with Local Feedback (AAAI 2025 · similarity 0.84)
- Graph-Based Algorithms for Diverse Similarity Search (ICML 2025 · similarity 0.83)
- Improving Consistency in Retrieval Augmented Systems With Group Similarity Rewards (NeurIPS 2026 · similarity 0.82)
- Ward: Provable RAG Dataset Inference via LLM Watermarks (ICLR 2025 · similarity 0.82)
- Relative Error Fair Clustering in the Weak-Strong Oracle Model (ICML 2025 · similarity 0.81)
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.
