Research map / Label-efficient and robust learning / Recognition and long-tail
Segment-driven Structural Induction and Semantic Alignment for Heterogeneous Tabular Representation
Woojun Jung, Susik Yoon
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 context-free attribute embedding (CAE) (ablation) | NAVI · Product header prediction | accuracy: 0.5996 -> 0.9995 | Improved |
| add context-free attribute embedding (CAE) (ablation) | NAVI · Movie value imputation | accuracy: 0.4799 -> 0.7077 | Improved |
| add context-free attribute embedding (CAE) (ablation) | NAVI · Movie row classification (XGBoost) | macro-F1: 0.265 -> 0.615 | Improved |
| add masked segment modeling instead of vanilla MLM (ablation) | NAVI · Product value imputation | accuracy: 0.7769 -> 0.7933 | Improved |
| add masked segment modeling instead of vanilla MLM (ablation) | NAVI · Movie row classification (XGBoost) | macro-F1: 0.467 -> 0.615 | Improved |
| add entropy-driven segment alignment (ESA) (ablation) | NAVI · Product value imputation | accuracy: 0.7977 -> 0.7933 | Hurt |
| add entropy-driven segment alignment (ESA) (ablation) | NAVI · Product row classification (XGBoost) | macro-F1: 0.909 -> 0.928 | Improved |
| add entropy-driven segment alignment (ESA) (ablation) | NAVI · Movie row classification (XGBoost) | macro-F1: 0.623 -> 0.615 | Hurt |
| use entropy-based routing instead of random routing in ESA (ablation) | NAVI · Product row classification (XGBoost) | macro-F1: 0.902 -> 0.928 | Improved |
| use entropy-based routing instead of random routing in ESA (ablation) | NAVI · Movie row classification (XGBoost) | macro-F1: 0.592 -> 0.615 | Improved |
| remove numerical columns (text-only) (ablation) | NAVI · Product row classification (XGBoost) | macro-F1: 0.928 -> 0.902 | Hurt |
| reduce textual coverage (partial-text) (ablation) | NAVI · Product row classification (XGBoost) | macro-F1: 0.928 -> 0.888 | Hurt |
Closest papers by meaning
- Multimodal Tabular Reasoning with Privileged Structured Information (NeurIPS 2025 · similarity 0.86)
- TRL-Bench: Standardizing Cross-Paradigm Representation-Level Evaluation of Tabular Encoders (NeurIPS 2026 · similarity 0.86)
- Enhancing Tabular Learners with Context-Aware Semantic Embeddings (NeurIPS 2026 · similarity 0.86)
- ConTextTab: A Semantics-Aware Tabular In-Context Learner (NeurIPS 2025 · similarity 0.86)
- Decoupling Skeleton and Flesh: Efficient Multimodal Table Reasoning with Disentangled Alignment and Structure-aware Guidance (ICML 2026 · similarity 0.85)
- UniTabE: A Universal Pretraining Protocol for Tabular Foundation Model in Data Science (ICLR 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.
