Research map / Large language models / Language, safety and interpretability
Deep Cascade Multi-Task Learning for Slot Filling in Online Shopping Assistant
Yu Gong, Xusheng Luo, Yu Zhu, Wenwu Ou, Zhao Li, Muhua Zhu, Kenny Q. Zhu, Lu Duan, Xi Chen
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 DCMTL with cascade and residual connections instead of Hierarchy Multi-task | ECSA slot filling · ECSA | F1: 0.4455 -> 0.5105 | Improved |
| Add residual connections (ablation) | DCMTL · ECSA | F1: 0.4840 -> 0.5105 | Improved |
| Add cascade connections (ablation) | DCMTL · ECSA | F1: 0.4633 -> 0.5105 | Improved |
| Use hierarchical multi-task learning rather than vanilla shared-encoder multi-task learning (ablation) | ECSA slot filling · ECSA | F1: 0.3965 -> 0.4455 | Improved |
| Provide the gold segment type as an input feature | Basic BiLSTM-CRF · ECSA | F1: 0.4302 -> 0.7950 | Improved |
| Provide the gold named entity type as an input feature | Basic BiLSTM-CRF · ECSA | F1: 0.4302 -> 0.8985 | Improved |
| Deploy DCMTL instead of online max-matching | Online shopping assistant · ECSA online queries | accuracy: 26.37% -> 60.53%; 130% improvement | Improved |
| Use DCMTL instead of Neural Sequence Chunking | ATIS slot filling · ATIS | F1: 0.9586 -> 0.9583; F1-0.03% | No clear effect |
| Use DCMTL instead of Sequential CNN | ATIS slot filling · ATIS | F1: 0.9561 -> 0.9583; F1+0.22% | Improved |
| Use cascade type 3, with segment tagging, named entity tagging, and slot filling in sequence, rather than a skipped connection (ablation) | DCMTL · ECSA | F1 | Improved |
Closest papers by meaning
- Shopping MMLU: A Massive Multi-Task Online Shopping Benchmark for Large Language Models (NeurIPS 2024 · similarity 0.84)
- EcomGPT: Instruction-Tuning Large Language Models with Chain-of-Task Tasks for E-commerce (AAAI 2024 · similarity 0.83)
- Introducing Visual Scenes and Reasoning: A More Realistic Benchmark for Spoken Language Understanding (AAAI 2026 · similarity 0.82)
- TA&AT: Enhancing Task-Oriented Dialog with Turn-Level Auxiliary Tasks and Action-Tree Based Scheduled Sampling (AAAI 2024 · similarity 0.81)
- Mix-Ecom: Towards Mixed-Type E-Commerce Dialogues with Complex Domain Rules (ICLR 2026 · similarity 0.81)
- Aligner²: Enhancing Joint Multiple Intent Detection and Slot Filling via Adjustive and Forced Cross-Task Alignment (AAAI 2024 · 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.
