EMNLP Findings2024
TARA: Token-level Attribute Relation Adaptation for Multi-Attribute Controllable Text Generation
Abstract
Multi-attribute controllable text generation (CTG) aims to generate fluent text satisfying multiple attributes, which is an important and challenging task. The majority of previous research on multi-attribute CTG has ignored the interrelations of attributes that affect the performance of text generation. Recently, several work considers the attribute relations by explicitly defining them as inhibtory. We argue that for multi-attribute CTG, the attribute relations are not fixed, which can be not only inhibtory but promotive as well. In this paper, we tackle the multi-attribute CTG problem by explicitly identifying the above attribute relations for the first time and propose TARA, which employs token-level attribute relation adaptation and representation to generate text with the balanced multi-attribute control. Experimental results on the benchmark dataset demonstrate the effectiveness of our proposed method.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Plug and Play Language Models: A Simple Approach to Controlled Text Generation cites The Curious Case of Neural Text Degeneration
- C³TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation cites Plug and Play Language Models: A Simple Approach to Controlled Text Generation
- C³TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation cites GeDi: Generative Discriminator Guided Sequence Generation
- C³TG: Conflict-aware, Composite, and Collaborative Controlled Text Generation cites Controllable Natural Language Generation with Contrastive Prefixes
- GeDi: Generative Discriminator Guided Sequence Generation cites Plug and Play Language Models: A Simple Approach to Controlled Text Generation
- GeDi: Generative Discriminator Guided Sequence Generation cites The Curious Case of Neural Text Degeneration
- Controllable Natural Language Generation with Contrastive Prefixes cites Plug and Play Language Models: A Simple Approach to Controlled Text Generation
- Controllable Natural Language Generation with Contrastive Prefixes cites GeDi: Generative Discriminator Guided Sequence Generation
Cited by top-tier papers1
Builds on12
- The Curious Case of Neural Text DegenerationAri Holtzman, Jan Buys, Li Du, Maxwell Forbes et al.ICLR 2020 · 4,112 citations
- Plug and Play Language Models: A Simple Approach to Controlled Text GenerationSumanth Dathathri, Andrea Madotto, Janice Lan, Jane Hung et al.ICLR 2020 · 1,166 citations
- GeDi: Generative Discriminator Guided Sequence GenerationBen Krause, Akhilesh Deepak Gotmare, Bryan McCann, Nitish Shirish Keskar et al.EMNLP Findings 2021 · 219 citations
- The Power of Scale for Parameter-Efficient Prompt TuningBrian Lester, Rami Al-Rfou, Noah ConstantEMNLP 2021 · 94 citations
- Controllable Natural Language Generation with Contrastive PrefixesJing Qian, Li Dong, Yelong Shen, Furu Wei et al.ACL Findings 2022 · 74 citations
Related papers
- Seen to Unseen: Exploring Compositional Generalization of Multi-Attribute Controllable Dialogue GenerationWeihao Zeng, Lulu Zhao, Keqing He, Ruotong Geng et al.ACL 2023 · 1 citation
- Attribute Alignment: Controlling Text Generation from Pre-trained Language ModelsDian Yu, Zhou Yu, Kenji SagaeEMNLP Findings 2021 · 26 citations
- Tailor: A Soft-Prompt-Based Approach to Attribute-Based Controlled Text GenerationKexin Yang, Dayiheng Liu, Wenqiang Lei, Baosong Yang et al.ACL 2023 · 25 citations
- Air-Decoding: Attribute Distribution Reconstruction for Decoding-Time Controllable Text GenerationTianqi Zhong, Quan Wang, Jingxuan Han, Yongdong Zhang et al.EMNLP 2023 · 7 citations
- Multi-Aspect Controllable Text Generation with Disentangled Counterfactual AugmentationYi Liu, Xiangyu Liu, Xiangrong Zhu, Wei HuACL 2024 · 6 citations
