Research map / Label-efficient and robust learning / Recognition and long-tail
Multi-Resolution Weak Supervision for Sequential Data
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 Dugong instead of Data Programming | End model on BAV · BAV | F1: +0.6 | Helped |
| Use Dugong instead of Traditional Supervision | End model on Interview · Interview | F1: +12.0 | Helped |
| Use Dugong instead of Data Programming | End model on Gait · Gait | F1: +5.1 | Helped |
| Use Dugong instead of Data Programming | End model on Shot · Shot | F1: +1.5 | Helped |
| Use Dugong instead of Traditional Supervision | End model on Basketball · Basketball | F1: +11.4 | Helped |
| Add sequential dependency modeling (ablation) | Dugong · Five sequential classification tasks (average) | F1: 10.2 F1 points | Helped |
| Share parameters across sources using the same logic at the same resolution (ablation) | Dugong · Five sequential classification tasks (average) | F1: 9.2 F1 points | Helped |
| Use a user-defined prior instead of a uniform prior (ablation) | Dugong · Five sequential classification tasks (average) | F1: 13.7 F1 points | Helped |
| Use a user-defined prior instead of a development-set-based prior (ablation) | Dugong · Five sequential classification tasks (average) | F1: 1.7 F1 points | Helped |
Closest papers by meaning
- AutoWS-Bench-101: Benchmarking Automated Weak Supervision with 100 Labels (NeurIPS 2022 · similarity 0.87)
- Training Complex Models with Multi-Task Weak Supervision (AAAI 2019 · similarity 0.87)
- WRENCH: A Comprehensive Benchmark for Weak Supervision (NeurIPS 2021 · similarity 0.87)
- Learning Dependency Structures for Weak Supervision Models (ICML 2019 · similarity 0.86)
- Unified Approach for Weakly Supervised Multicalibration (NeurIPS 2026 · similarity 0.86)
- Weak Supervision Performance Evaluation via Partial Identification (NeurIPS 2024 · similarity 0.85)
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.