Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Meta Multi-Task Learning for Sequence Modeling
Junkun Chen, Xipeng Qiu, Pengfei Liu, Xuanjing Huang
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 |
|---|---|---|---|
| Share a Meta-LSTM that generates task-specific, position-dependent composition parameters instead of using feature-level shared-private layers | Multi-task sequence modeling · Classification AVG | accuracy: +2.2% to best competitor Multi-task model | Improved |
| Use Meta-LSTM-generated dynamic composition parameters instead of a standard LSTM | Single-task text classification · 16 classification datasets | accuracy: 3.2% | Improved |
| Transfer the trained shared Meta-LSTM and freeze its parameters on a new task | Meta-MTL · Classification AVG | accuracy: 3.1% | Improved |
| Vary the meta-vector size z from 20 to 60 (ablation) | Meta-MTL · 16 classification datasets | average accuracy: less than 0.8% | No clear effect |
| Use Meta-LSTM-MTL+CRF instead of LSTM-SSP-MTL+CRF | Sequence tagging · CoNLL 2000 | F1: 94.32 -> 95.11 | Improved |
| Use Meta-LSTM-MTL+CRF instead of LSTM-SSP-MTL+CRF | Sequence tagging · CoNLL 2003 | F1: 90.38 -> 90.72 | Improved |
| Use Meta-LSTM-MTL+CRF instead of LSTM-SSP-MTL+CRF | Sequence tagging · WSJ | accuracy: 97.23 -> 97.45 | Improved |
Closest papers by meaning
- Compositional generalization through meta sequence-to-sequence learning (NeurIPS 2019 · similarity 0.87)
- Compositional Task Representations for Large Language Models (ICLR 2023 · similarity 0.87)
- Learning Multi-Task Communication with Message Passing for Sequence Learning (AAAI 2019 · similarity 0.86)
- Any-Way Meta Learning (AAAI 2024 · similarity 0.85)
- Consistency of Compositional Generalization Across Multiple Levels (AAAI 2025 · similarity 0.84)
- Context-Aware Meta-Learning (ICLR 2024 · similarity 0.84)
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.
