Reasoning Graph Enhanced Exemplars Retrieval for In-Context Learning
This addresses the challenge of exemplar selection for researchers and practitioners using in-context learning, though it is incremental as it builds on existing retrieval methods by adding structural similarity.
The paper tackles the problem of selecting high-quality exemplars for in-context learning in large language models by proposing a method that uses reasoning graphs to capture logical connections between steps, resulting in improved performance on math and logic reasoning tasks over state-of-the-art retrieval-based approaches.
Large language models (LLMs) have exhibited remarkable few-shot learning capabilities and unified the paradigm of NLP tasks through the in-context learning (ICL) technique. Despite the success of ICL, the quality of the exemplar demonstrations can significantly influence the LLM's performance. Existing exemplar selection methods mainly focus on the semantic similarity between queries and candidate exemplars. On the other hand, the logical connections between reasoning steps can be beneficial to depict the problem-solving process as well. In this paper, we proposes a novel method named Reasoning Graph-enhanced Exemplar Retrieval (RGER). RGER first quires LLM to generate an initial response, then expresses intermediate problem-solving steps to a graph structure. After that, it employs graph kernel to select exemplars with semantic and structural similarity. Extensive experiments demonstrate the structural relationship is helpful to the alignment of queries and candidate exemplars. The efficacy of RGER on math and logit reasoning tasks showcases its superiority over state-of-the-art retrieval-based approaches. Our code is released at https://github.com/Yukang-Lin/RGER.