AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Language Models Can Improve Event Prediction by Few-Shot Abductive Reasoning

Xiaoming Shi, Siqiao Xue, Kangrui Wang, Fan Zhou, James Y. Zhang, JUN ZHOU, Chenhao Tan, Hongyuan Mei

NeurIPS 2023 · Large language models › Language, safety and interpretability · approach: Faithfulness of LLM explanations · 12 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LLM abductive reasoning and evidence ranking (LAMP) on top of base event modelsevent sequence models (NHP, KE, DyRep, ANHP) · GDELT predicate-object predictionmean rank: around 40 -> around 20Improved
add LAMP to base event modelsevent sequence models · ICEWS object predictionmean rank: around 90 -> about 20Improved
use Llama-2-chat 13B instead of GPT-3.5 as the LLM (ablation)LAMP · GDELT predicate/object predictionmean rankNo clear effect
use GPT-3-davinci instead of GPT-3.5 or Llama (ablation)LAMP · GDELTmean rankHurt
increase retrieved evidence events from 5 to 10, then beyond 10 (ablation)LAMP · GDELTmean rankMixed
use more demonstrations in the prompt (ablation)LAMP · GDELTmean rankImproved
replace half of the demonstrations (P1 to P2) (ablation)LAMP · GDELTmean rankNo clear effect
use LLM-generated causes for retrieval instead of random, recent, same-predicate, same-object or SBERT retrieval (ablation)LAMP · GDELTmean rankImproved
use similarity threshold instead of fixed number of retrieved events (ablation)LAMP · GDELTmean rankNo clear effect
use edit distance instead of SBERT cosine similarity for retrieval (ablation)LAMP · GDELTmean rankHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy