27 Papers
35 Citations
Xi Ye is an academic researcher from University of Texas at Austin. The author has contributed to research in topics: Computer science & Natural language. The author has an hindex of 4, co-authored 16 publications. Previous affiliations of Xi Ye include Tsinghua University.
Chat about Author
Papers
Proceedings Article
The Unreliability of Explanations in Few-shot Prompting for Textual Reasoning
Xi Ye,Gregory Christopher Durrett +1 more
- 06 May 2022
TL;DR: Analysis in three settings shows that explanations judged by humans to be good—logically consistent with the input and the prediction—more likely cooccur with accurate predictions, and trains calibrators using automatically extracted scores that assess the reliability of explanations to improve performance post-hoc.
111
•Posted Content
RnG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering
TL;DR: RnG-KBQA as mentioned in this paper uses a contrastive ranker to rank a set of candidate logical forms obtained by searching over the knowledge graph, and then introduces a tailored generation model conditioned on the question and the top-ranked candidates to compose the final logical form.
36
Sketch-Driven Regular Expression Generation from Natural Language and Examples
TL;DR: This paper proposed a system for converting natural language descriptions into regular expressions (regexes) using regular expressions and regular expressions can be converted to natural language sentences, but typically deal with short, formulaic text and can only produce simple regular expressions.
Benchmarking Multimodal Regex Synthesis with Complex Structures.
Xi Ye,Qiaochu Chen,Isil Dillig,Greg Durrett +3 more
- 01 Jul 2020
TL;DR: This work introduces StructuredRegex, a new regex synthesis dataset differing from prior ones in three aspects, to obtain structurally complex and realistic regexes, using a probabilistic grammar with pre-defined macros observed from real-world StackOverflow posts.
•Posted Content
Sketch-Driven Regular Expression Generation from Natural Language and Examples
TL;DR: This work presents a framework for regex synthesis in this setting where both natural language (NL) and examples are available, and achieves state-of-the-art performance on the prior datasets and solves 57% of the real-world dataset, which existing neural systems completely fail on.
14