2000 character limit reached
Improving Open-Ended Text Generation via Adaptive Decoding (2402.18223v2)
Published 28 Feb 2024 in cs.CL
Abstract: Current LLMs decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a mechanism that dynamically empowers LLMs to ascertain a sensible candidate set during generation. Specifically, we introduce an entropy-based metric called confidence and conceptualize determining the optimal candidate set as a confidence-increasing process. The rationality of including a token in the candidate set is assessed by leveraging the increment of confidence. Experimental results reveal that our method balances diversity and coherence well. The human evaluation shows that our method can generate human-preferred text. Additionally, our method can potentially improve the reasoning ability of LLMs.
- Mirostat: a neural text decoding algorithm that directly controls perplexity. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net, 2021.
- Language models are few-shot learners. In Larochelle, H., Ranzato, M., Hadsell, R., Balcan, M., and Lin, H. (eds.), Advances in Neural Information Processing Systems, volume 33, pp. 1877–1901. Curran Associates, Inc., 2020.
- Hierarchical neural story generation. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 889–898, Melbourne, Australia, July 2018. Association for Computational Linguistics. doi: 10.18653/v1/P18-1082.
- SimCSE: Simple contrastive learning of sentence embeddings. In Moens, M.-F., Huang, X., Specia, L., and Yih, S. W.-t. (eds.), Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp. 6894–6910, Online and Punta Cana, Dominican Republic, November 2021. Association for Computational Linguistics. doi: 10.18653/v1/2021.emnlp-main.552.
- Truncation sampling as language model desmoothing. In Findings of the Association for Computational Linguistics: EMNLP 2022, pp. 3414–3427, Abu Dhabi, United Arab Emirates, December 2022. Association for Computational Linguistics.
- The curious case of neural text degeneration. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net, 2020.
- Don’t say that! making inconsistent dialogue unlikely with unlikelihood training. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp. 4715–4728, Online, July 2020. Association for Computational Linguistics. doi: 10.18653/v1/2020.acl-main.428.
- On the efficacy of sampling adapters. In Rogers, A., Boyd-Graber, J., and Okazaki, N. (eds.), Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 1437–1455, Toronto, Canada, July 2023a. Association for Computational Linguistics. doi: 10.18653/v1/2023.acl-long.80.
- Locally typical sampling. Transactions of the Association for Computational Linguistics, 11:102–121, 2023b.
- Pointer sentinel mixture models, 2016.
- A corpus and cloze evaluation for deeper understanding of commonsense stories. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 839–849, San Diego, California, June 2016. Association for Computational Linguistics. doi: 10.18653/v1/N16-1098.
- Mauve: Measuring the gap between neural text and human text using divergence frontiers. In Ranzato, M., Beygelzimer, A., Dauphin, Y., Liang, P., and Vaughan, J. W. (eds.), Advances in Neural Information Processing Systems, volume 34, pp. 4816–4828. Curran Associates, Inc., 2021.
- Language models are unsupervised multitask learners. OpenAI blog, 1(8):9, 2019.
- Emo: Earth mover distance optimization for auto-regressive language modeling, 2023.
- A contrastive framework for neural text generation. In Koyejo, S., Mohamed, S., Agarwal, A., Belgrave, D., Cho, K., and Oh, A. (eds.), Advances in Neural Information Processing Systems, volume 35, pp. 21548–21561. Curran Associates, Inc., 2022.
- Lamda: Language models for dialog applications. arXiv preprint arXiv:2201.08239, 2022.
- Llama 2: Open foundation and fine-tuned chat models, 2023.
- Neural text generation with unlikelihood training, 2019.
- Learning to break the loop: Analyzing and mitigating repetitions for neural text generation. In Koyejo, S., Mohamed, S., Agarwal, A., Belgrave, D., Cho, K., and Oh, A. (eds.), Advances in Neural Information Processing Systems, volume 35, pp. 3082–3095. Curran Associates, Inc., 2022.
- Look-back decoding for open-ended text generation. In Bouamor, H., Pino, J., and Bali, K. (eds.), Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 1039–1050, Singapore, December 2023. Association for Computational Linguistics. doi: 10.18653/v1/2023.emnlp-main.66.
- Penalty decoding: Well suppress the self-reinforcement effect in open-ended text generation. In Bouamor, H., Pino, J., and Bali, K. (eds.), Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pp. 1218–1228, Singapore, December 2023. Association for Computational Linguistics. doi: 10.18653/v1/2023.emnlp-main.78.
Collections
Sign up for free to add this paper to one or more collections.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
