CASE: Commonsense-Augmented Score with an Expanded Answer Space
This work addresses improving zero-shot QA accuracy for NLP applications, though it is incremental as it builds on prior methods like answer space expansion.
The authors tackled the limitation of basic language model scores in multiple-choice QA by proposing CASE, a commonsense-augmented scoring method with dynamic word weighting, which outperformed baselines on 5 commonsense benchmarks when combined with answer space expansion.
LLMs have demonstrated impressive zero-shot performance on NLP tasks thanks to the knowledge they acquired in their training. In multiple-choice QA tasks, the LM probabilities are used as an imperfect measure of the plausibility of each answer choice. One of the major limitations of the basic score is that it treats all words as equally important. We propose CASE, a Commonsense-Augmented Score with an Expanded Answer Space. CASE addresses this limitation by assigning importance weights for individual words based on their semantic relations to other words in the input. The dynamic weighting approach outperforms basic LM scores, not only because it reduces noise from unimportant words, but also because it informs the model of implicit commonsense knowledge that may be useful for answering the question. We then also follow prior work in expanding the answer space by generating lexically-divergent answers that are conceptually-similar to the choices. When combined with answer space expansion, our method outperforms strong baselines on 5 commonsense benchmarks. We further show these two approaches are complementary and may be especially beneficial when using smaller LMs.