Plan, Attend, Generate: Planning for Sequence-to-Sequence Models

2017-11-28 · NeurIPS 2017 12 · Francis Dutil, Caglar Gulcehre, Adam Trischler, Yoshua Bengio

We investigate the integration of a planning mechanism into sequence-to-sequence models using attention. We develop a model which can plan ahead in the future when it computes its alignments between input and output sequences, constructing a matrix of proposed future alignments and a commitment vector that governs whether to follow or recompute the plan. This mechanism is inspired by the recently proposed strategic attentive reader and writer (STRAW) model for Reinforcement Learning. Our proposed model is end-to-end trainable using primarily differentiable operations. We show that it outperforms a strong baseline on character-level translation tasks from WMT'15, the algorithmic task of finding Eulerian circuits of graphs, and question generation from the text. Our analysis demonstrates that the model computes qualitatively intuitive alignments, converges faster than the baselines, and achieves superior performance with fewer parameters.

📄 PDF Abstract BibTeX arXiv:1711.10462

Code (1)

nyu-dl/dl4mt-cdec 공식 구현

Tasks

Question Generation Question-Generation Reinforcement Learning Reinforcement Learning (RL)Translation

Plan, Attend, Generate: Character-Level Neural Machine Translation with Planning

2017-08-01 · WS 2017 8 · Caglar Gulcehre, Francis Dutil, Adam Trischler, Yoshua Bengio

We investigate the integration of a planning mechanism into an encoder-decoder architecture with attention. We develop a model that can plan ahead when it computes alignments between the source and target sequences not o…

Decoder Hierarchical Reinforcement Learning Machine Translation Reinforcement Learning+3

Plan, Attend, Generate: Character-level Neural Machine Translation with Planning in the Decoder

2017-06-13 · Caglar Gulcehre, Francis Dutil, Adam Trischler, Yoshua Bengio

We investigate the integration of a planning mechanism into an encoder-decoder architecture with an explicit alignment for character-level machine translation. We develop a model that plans ahead when it computes alignme…

Decoder Machine Translation Translation

GraphPlan: Story Generation by Planning with Event Graph

2021-02-05 · INLG (ACL) 2021 8 · Hong Chen, Raphael Shu, Hiroya Takamura, Hideki Nakayama

Story generation is a task that aims to automatically produce multiple sentences to make up a meaningful story. This task is challenging because it requires high-level understanding of semantic meaning of sentences and c…

Story Generation Text Generation

Large Language Models Can Take False First Steps at Inference-time Planning

2026-02-03 · Haijiang Yan, Jian-Qiao Zhu, Adam Sanborn arxiv

Large language models (LLMs) have been shown to acquire sequence-level planning abilities during training, yet their planning behavior exhibited at inference time often appears short-sighted and inconsistent with these c…

Plan-CVAE: A Planning-based Conditional Variational Autoencoder for Story Generation

2020-10-01 · CCL 2020 10 · Lin Wang, Juntao Li, Rui Yan, Dongyan Zhao

Story generation is a challenging task of automatically creating natural languages to describe a sequence of events, which requires outputting text with not only a consistent topic but also novel wordings. Although many …

Diversity Story Generation