Papers with explanation graph generation

2 papers
MIDGARD: Self-Consistency Using Minimum Description Length for Structured Commonsense Reasoning (2024.acl-long)

Copied to clipboard

Challenge: Existing approaches to generate a reasoning graph from natural language input suffer from error propagation due to autoregressive nature and single-pass-based decoding.
Approach: They propose a method that uses minimum description length to identify consistent properties among different graph samples generated by large language models.
Outcome: The proposed method outperforms previous approaches for generating reasoning graphs from natural language input using large language models.
ExplaGraphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning (2021.emnlp-main)

Copied to clipboard

Challenge: Current commonsense-reasoning tasks are discriminative in nature, where a model answers a multiple-choice question for a certain context.
Approach: They propose a generative task that generates a commonsense-augmented graph for stance prediction by using a create-verify-and-refine graph collection framework.
Outcome: The proposed model is able to generate a graph that serves as non-trivial, complete, and unambiguous explanation for the predicted stance.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations