Introduction

This chapter covers how to make completions more reliable, as well as how to implement checks to ensure that outputs are reliable.

To a certain extent, most of the previous techniques covered have to do with improving the completion accuracy, and thus reliability, in particular self-consistency. However, several other techniques can be used to improve reliability, beyond basic prompting strategies.

LLMs are more reliable than we might expect at interpreting what a prompt is trying to say when responding to misspelled, badly phrased, or even actively misleading prompts. Despite this ability, they still exhibit various problems including hallucinations, flawed explanations with CoT prompting methods, and multiple biases including majority label bias, recency bias, and common token bias. Additionally, Zero-Shot CoT can be particularly biased when dealing with sensitive topics.

Common solutions to some of these problems include calibrators to remove a priori biases, and verifiers to score completions, as well as promoting diversity in completions.

🟦 Calibrating LLMs

🟢 Prompt Debiasing

🟦 Prompt Ensembling

🟦 LLM Self-Evaluation

🟦 Math

Footnotes

1. Wang, X., Wei, J., Schuurmans, D., Le, Q., Chi, E., Narang, S., Chowdhery, A., & Zhou, D. (2022). Self-Consistency Improves Chain of Thought Reasoning in Language Models. ↩

2. Webson, A., Loo, A. M., Yu, Q., & Pavlick, E. (2023). Are Language Models Worse than Humans at Following Prompts? It’s Complicated. arXiv:2301.07085v1 [Cs.CL]. ↩

3. Ye, X., & Durrett, G. (2022). The Unreliability of Explanations in Few-shot Prompting for Textual Reasoning. ↩ ↩²

4. Zhao, T. Z., Wallace, E., Feng, S., Klein, D., & Singh, S. (2021). Calibrate Before Use: Improving Few-Shot Performance of Language Models. ↩

5. Shaikh, O., Zhang, H., Held, W., Bernstein, M., & Yang, D. (2022). On Second Thought, Let’s Not Think Step by Step! Bias and Toxicity in Zero-Shot Reasoning. ↩