Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386875188> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W4386875188 abstract "Knowledge editing aims to change language models' performance on several special cases (i.e., editing scope) by infusing the corresponding expected knowledge into them. With the recent advancements in large language models (LLMs), knowledge editing has been shown as a promising technique to adapt LLMs to new knowledge without retraining from scratch. However, most of the previous studies neglect the multi-lingual nature of some main-stream LLMs (e.g., LLaMA, ChatGPT and GPT-4), and typically focus on monolingual scenarios, where LLMs are edited and evaluated in the same language. As a result, it is still unknown the effect of source language editing on a different target language. In this paper, we aim to figure out this cross-lingual effect in knowledge editing. Specifically, we first collect a large-scale cross-lingual synthetic dataset by translating ZsRE from English to Chinese. Then, we conduct English editing on various knowledge editing methods covering different paradigms, and evaluate their performance in Chinese, and vice versa. To give deeper analyses of the cross-lingual effect, the evaluation includes four aspects, i.e., reliability, generality, locality and portability. Furthermore, we analyze the inconsistent behaviors of the edited models and discuss their specific challenges." @default.
- W4386875188 created "2023-09-20" @default.
- W4386875188 creator A5016730781 @default.
- W4386875188 creator A5062115445 @default.
- W4386875188 creator A5067510740 @default.
- W4386875188 creator A5071554124 @default.
- W4386875188 creator A5083841358 @default.
- W4386875188 date "2023-09-16" @default.
- W4386875188 modified "2023-09-27" @default.
- W4386875188 title "Cross-Lingual Knowledge Editing in Large Language Models" @default.
- W4386875188 doi "https://doi.org/10.48550/arxiv.2309.08952" @default.
- W4386875188 hasPublicationYear "2023" @default.
- W4386875188 type Work @default.
- W4386875188 citedByCount "0" @default.
- W4386875188 crossrefType "posted-content" @default.
- W4386875188 hasAuthorship W4386875188A5016730781 @default.
- W4386875188 hasAuthorship W4386875188A5062115445 @default.
- W4386875188 hasAuthorship W4386875188A5067510740 @default.
- W4386875188 hasAuthorship W4386875188A5071554124 @default.
- W4386875188 hasAuthorship W4386875188A5083841358 @default.
- W4386875188 hasBestOaLocation W43868751881 @default.
- W4386875188 hasConcept C154945302 @default.
- W4386875188 hasConcept C15744967 @default.
- W4386875188 hasConcept C199360897 @default.
- W4386875188 hasConcept C204321447 @default.
- W4386875188 hasConcept C2778012447 @default.
- W4386875188 hasConcept C2780767217 @default.
- W4386875188 hasConcept C41008148 @default.
- W4386875188 hasConcept C542102704 @default.
- W4386875188 hasConcept C63000827 @default.
- W4386875188 hasConceptScore W4386875188C154945302 @default.
- W4386875188 hasConceptScore W4386875188C15744967 @default.
- W4386875188 hasConceptScore W4386875188C199360897 @default.
- W4386875188 hasConceptScore W4386875188C204321447 @default.
- W4386875188 hasConceptScore W4386875188C2778012447 @default.
- W4386875188 hasConceptScore W4386875188C2780767217 @default.
- W4386875188 hasConceptScore W4386875188C41008148 @default.
- W4386875188 hasConceptScore W4386875188C542102704 @default.
- W4386875188 hasConceptScore W4386875188C63000827 @default.
- W4386875188 hasLocation W43868751881 @default.
- W4386875188 hasOpenAccess W4386875188 @default.
- W4386875188 hasPrimaryLocation W43868751881 @default.
- W4386875188 hasRelatedWork W110609005 @default.
- W4386875188 hasRelatedWork W1517479908 @default.
- W4386875188 hasRelatedWork W2077030552 @default.
- W4386875188 hasRelatedWork W2120994104 @default.
- W4386875188 hasRelatedWork W2211967836 @default.
- W4386875188 hasRelatedWork W2365235480 @default.
- W4386875188 hasRelatedWork W2377544373 @default.
- W4386875188 hasRelatedWork W2784803070 @default.
- W4386875188 hasRelatedWork W4250353049 @default.
- W4386875188 hasRelatedWork W2794878329 @default.
- W4386875188 isParatext "false" @default.
- W4386875188 isRetracted "false" @default.
- W4386875188 workType "article" @default.