Research map / Time series / Foundation models for time series
TGCD: A Framework for Generalized Category Discovery in Time-Series Data
Chandan Gautam, Lew Choon Hean, Ankit Das, Xiaoli Li, Savitha Ramasamy
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 |
|---|---|---|---|
| add KUTD (temporal discriminability) loss (ablation) | TGCD · CharacterTrajectories | accuracy (unknown): 76.32 -> 80.85 | Improved |
| add STeSD segment dropout loss (ablation) | TGCD · CharacterTrajectories | accuracy (unknown): 74.70 -> 80.85 | Improved |
| add STeSD and KUTD to margin-aware CE (ablation) | TGCD · CharacterTrajectories | accuracy (all): 66.85 -> 73.01 | Improved |
| add temporal losses (ablation) | TGCD · CharacterTrajectories | accuracy (known): 94.45 -> 95.26 | No clear effect |
| use MOMENT with fine-tuning instead of frozen (ablation) | TGCD · CharacterTrajectories | accuracy (all): 11.47 -> 73.01 | Improved |
| use MOMENT instead of TS-TCC backbone (ablation) | TGCD · CharacterTrajectories | accuracy (unknown): 66.49 -> 80.85 | Improved |
| TGCD vs best baseline overall | TGCD · six time-series datasets | accuracy (all): 73.01 vs 69.23 (ORCA) on CharacterTrajectories | Improved |
| raise labeled sample ratio beyond 10-30% (ablation) | TGCD · CharacterTrajectories | accuracy (all, unknown) | Hurt |
| raise known category ratio (ablation) | TGCD · CharacterTrajectories | accuracy (all, unknown) | Improved |
Closest papers by meaning
- Solving the Catastrophic Forgetting Problem in Generalized Category Discovery (CVPR 2024 · similarity 0.88)
- DebGCD: Debiased Learning with Distribution Guidance for Generalized Category Discovery (ICLR 2025 · similarity 0.88)
- Prior-Constrained Association Learning for Fine-Grained Generalized Category Discovery (AAAI 2025 · similarity 0.88)
- HIDISC: A Hyperbolic Framework for Domain Generalization with Generalized Category Discovery (NeurIPS 2025 · similarity 0.88)
- Transfer and Alignment Network for Generalized Category Discovery (AAAI 2024 · similarity 0.87)
- Continual Generalized Category Discovery: Learning and Forgetting from a Bayesian Perspective (ICML 2025 · similarity 0.87)
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.
