Retrieval

We study how LLMs handle irrelevant information and find they struggle with content that is semantically related but ultimately not pertinent, highlighting the limitations of current systems in filtering out such distractions.

We proposes EASYTOOL, a method that simplifies tool documentation into concise instructions, improving tool use by language models.

We propose a multilingual idiom KB (IdiomKB) developed using LLMs to facilitate better idiomatic translation by smaller models by retrieving idioms’ figurative meanings.

We present the first comprehensive and controlled investigation into the behavior of large language models when encountering knowledge conflicts.

Fact verification guided iterative factual error correction without the supervision from corretion.

Interpretable fact verification with phrasal decomposition and logic regularization.