Abstract
Controlled Text Generation (CTG) steers the generation of continuations of a given context (prompt) by a Large Language Model (LLM) towards texts possessing a given attribute (e.g., topic, sentiment). In this paper we view CTG as a Continual Learning problem: how to learn at every step to steer next-word generation, without having to wait for end-of-sentence. This continual view is useful for online applications such as CTG for speech, where end-of-sentence is often uncertain. We depart from an existing model, the Plug-and-Play language models (PPLM), which perturbs the context at each step to better predict next-words that posses the desired attribute. While PPLM is intricate and has many hyper-parameters, we provide a proof that the PPLM objective function can be reduced to a Continual Reinforcement Learning (CRL) reward function, thereby simplifying PPLM and endowing it with a better understood learning framework. Subsequently, we present, the first of its kind, CTG algorithm that is fully based on CRL and exhibit promising empirical results.- Anthology ID:
- 2024.lrec-main.343
- Volume:
- Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
- Month:
- May
- Year:
- 2024
- Address:
- Torino, Italia
- Editors:
- Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue
- Venues:
- LREC | COLING
- SIG:
- Publisher:
- ELRA and ICCL
- Note:
- Pages:
- 3881–3889
- Language:
- URL:
- https://aclanthology.org/2024.lrec-main.343
- DOI:
- Cite (ACL):
- Velizar Shulev and Khalil Sima’an. 2024. Continual Reinforcement Learning for Controlled Text Generation. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 3881–3889, Torino, Italia. ELRA and ICCL.
- Cite (Informal):
- Continual Reinforcement Learning for Controlled Text Generation (Shulev & Sima’an, LREC-COLING 2024)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2024.lrec-main.343.pdf