Unlocking Anticipatory Text Generation: A Constrained Approach for Large Language Models Decoding

arXiv:2312.06149v2 Announce Type: replace
Abstract: Large Language Models (LLMs) have demonstrated a powerful ability for text generation. However, achieving optimal results with a given prompt or instruction can be challenging, especially for billion-sized models. Additionally, undesired behaviors such as toxicity or hallucinations can manifest. While much larger models (e.g., ChatGPT) may demonstrate strength in mitigating these issues, there is still no guarantee of complete prevention. In this work, we propose formalizing text generation as a future-constrained generation problem to minimize …

abstract arxiv billion cs.ai cs.cl decoding hallucinations language language models large language large language models larger models llms manifest prompt text text generation toxicity type