SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2209.04761 · 2022

Testing Pre-trained Language Models' Understanding of Distributivity via Causal Mediation Analysis

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

To what extent do pre-trained language models grasp semantic knowledge regarding the phenomenon of distributivity? In this paper, we introduce DistNLI, a new diagnostic dataset for natural language inference that targets the semantic difference arising from distributivity, and employ the causal mediation analysis framework to quantify the model behavior and explore the underlying mechanism in this semantically-related task. We find that the extent of models' understanding is associated with model size and vocabulary size. We also provide insights into how models encode such high-level semantic knowledge.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2209.04761")
get_code_for_paper("2209.04761")
have("2209.04761")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2209.04761.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2209.04761.svg)](https://syntology.ai/paper/2209.04761)

Connect an agent — have() is free.