Harnessing Black-Box Control to Boost Commonsense in LM’s Generation

Yufei Tian, Felix Zhang, Nanyun Peng


Abstract
Large language models (LLMs) such as GPT-3 have demonstrated a strong capability to generate coherent and contextually relevant text. However, amidst their successes, a crucial issue persists: their generated outputs still lack commonsense at times. Moreover, fine-tuning the entire LLM towards more commonsensical outputs is computationally expensive if not infeasible. In this paper, we present a computation-efficient framework that steers a frozen Pre-Trained Language Model (PTLM) towards more commonsensical generation (i.e., producing a plausible output that incorporates a list of concepts in a meaningful way). Specifically, we first construct a reference-free evaluator that assigns a sentence with a commonsensical score by grounding the sentence to a dynamic commonsense knowledge base from four different relational aspects. We then use the scorer as the oracle for commonsense knowledge, and extend the controllable generation method called NADO to train an auxiliary head that guides a fixed PTLM to better satisfy the oracle. We test our framework on a series of GPT-2-, Flan-T5-, and Alpaca-based language models (LMs) on two constrained concept-to-sentence benchmarks. Human evaluation results demonstrate that our method consistently leads to the most commonsensical outputs.
Anthology ID:
2023.emnlp-main.329
Volume:
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5417–5432
Language:
URL:
https://aclanthology.org/2023.emnlp-main.329
DOI:
10.18653/v1/2023.emnlp-main.329
Bibkey:
Cite (ACL):
Yufei Tian, Felix Zhang, and Nanyun Peng. 2023. Harnessing Black-Box Control to Boost Commonsense in LM’s Generation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 5417–5432, Singapore. Association for Computational Linguistics.
Cite (Informal):
Harnessing Black-Box Control to Boost Commonsense in LM’s Generation (Tian et al., EMNLP 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/dois-2013-emnlp/2023.emnlp-main.329.pdf
Video:
 https://preview.aclanthology.org/dois-2013-emnlp/2023.emnlp-main.329.mp4