Surface Form Competition: Why the Highest Probability Answer Isn't Always Right

by   Ari Holtzman, et al.

Large language models have shown promising results in zero-shot settings (Brown et al.,2020; Radford et al., 2019). For example, they can perform multiple choice tasks simply by conditioning on a question and selecting the answer with the highest probability. However, ranking by string probability can be problematic due to surface form competition-wherein different surface forms compete for probability mass, even if they represent the same underlying concept, e.g. "computer" and "PC." Since probability mass is finite, this lowers the probability of the correct answer, due to competition from other strings that are valid answers (but not one of the multiple choice options). We introduce Domain Conditional Pointwise Mutual Information, an alternative scoring function that directly compensates for surface form competition by simply reweighing each option according to a term that is proportional to its a priori likelihood within the context of the specific zero-shot task. It achieves consistent gains in zero-shot performance over both calibrated (Zhao et al., 2021) and uncalibrated scoring functions on all GPT-2 and GPT-3 models over a variety of multiple choice datasets.


page 4

page 5

page 6

page 7

page 8

page 9

page 10

page 11


Evaluating Prompts Across Multiple Choice Tasks In a Zero-Shot Setting

Large language models have shown that impressive zero-shot performance c...

Prompt Consistency for Zero-Shot Task Generalization

One of the most impressive results of recent NLP history is the ability ...

Large Language Models are Zero-Shot Clinical Information Extractors

We show that large language models, such as GPT-3, perform well at zero-...

Billions of Parameters Are Worth More Than In-domain Training Data: A case study in the Legal Case Entailment Task

Recent work has shown that language models scaled to billions of paramet...

Zero-shot Neural Retrieval via Domain-targeted Synthetic Query Generation

Deep neural scoring models have recently been shown to improve ranking q...

A Zero-Shot Classification Approach for a Word-Guessing Challenge

The Taboo Challenge competition, a task based on the well-known Taboo ga...

Language Models (Mostly) Know What They Know

We study whether language models can evaluate the validity of their own ...