Research map / Time series / Causal and generative time series
Convolution Kernels for Discriminative Learning from Streaming Text
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 text ◦ time convolution kernel instead of text+time fusion kernel | GP · Synthetic classification | accuracy: 0.57 ± 0.12 -> 0.80 ± 0.16 | Helped |
| use text ◦ time kernel instead of text-only kernel | GP · Synthetic classification | accuracy: 0.62 ± 0.16 -> 0.80 ± 0.16 | Helped |
| use text ◦ time kernel instead of time kernel | GP · Synthetic regression | MSE: 0.15 ± 0.07 -> 0.05 ± 0.03 | Helped |
| use text ◦ time kernel instead of text+time | GP · Synthetic Poisson regression | log likelihood: -37.01 ± 8.66 -> -31.00 ± 5.63 | Helped |
| use simple kernels (text, time, text+time) instead of mean/majority baseline | GP · Synthetic data | accuracy / MSE / LL: 0.44 ± 0.04 -> 0.57 ± 0.12 | No clear effect |
| use text ◦ time kernel instead of time kernel | GP · Ottawa+Ferguson rumour classification | accuracy: 0.740 ± 0.015 -> 0.748 ± 0.009 | Helped |
| use text ◦ time kernel instead of majority baseline | GP · Ottawa+Ferguson rumour classification | accuracy: 0.701 ± 0.021 -> 0.748 ± 0.009 | Helped |
| use text ◦ time kernel instead of text+time | GP · Ottawa+Ferguson rumour classification | accuracy: 0.738 ± 0.011 -> 0.748 ± 0.009 | Helped |
| use text-only kernel instead of majority baseline | GP · Ottawa+Ferguson rumour classification | accuracy: 0.701 ± 0.021 -> 0.678 ± 0.025 | Hurt |
| use LGCP text ◦ time instead of LGCP ICM+TXT | LGCP · Ferguson rumour counts | MSE: 2.31±7.80 -> 2.22±7.12 | Helped |
| use LGCP text ◦ time instead of LGCP ICM+TXT | LGCP · Ferguson rumour counts | log likelihood: -14.6±10.8 -> -14.2±8.7 | Helped |
| add rumour text kernel (TXT) to LGCP | LGCP · Ferguson rumour counts | MSE: 3.44±9.99 -> 2.32±7.06 | Helped |
Closest papers by meaning
- Kernel Representation Learning for Time Sequence: Algorithm, Theory, and Applications (AAAI 2025 · similarity 0.83)
- Time Series Generation Under Data Scarcity: A Unified Generative Modeling Approach (NeurIPS 2025 · similarity 0.82)
- TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting (ICLR 2024 · similarity 0.82)
- Boosting Transferability and Discriminability for Time Series Domain Adaptation (NeurIPS 2024 · similarity 0.82)
- Large Pre-trained time series models for cross-domain Time series analysis tasks (NeurIPS 2024 · similarity 0.82)
- Generating Financial Time Series by Matching Random Convolutional Features (NeurIPS 2026 · similarity 0.82)
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.