LLM Digest

Story

arxiv_cs_cl · Nov 12, 2026 · paper

Source brief

Shortcuts or Memorisation? Revisiting Data Corruption in Natural Language Understanding with Large Language Models

arxiv.orgNov 12, 2026
original source linked

In brief

Pre-trained encoders fine-tuned on natural language inference (NLI) and other language understanding tasks keep much of their accuracy after the input is perturbed, e.g. by removing all nouns or shuffling the words. D...

Read the original at arxiv.org →Open in live feed Read that day’s brief

Shortcuts or Memorisation? Revisiting Data Corruption in Natural Language Understanding with Large Language Models

Earlier in this thread 4 items

LEMMO: A Large Electromagnetic Model leveraging Pretrained LLMs for Million-Sample Signal Understanding

Understanding Autonomous Driving Datasets by Describing Differences between Image Subsets in Natural Language

Natural Synthesis: Outperforming Reactive Synthesis Tools with Large Reasoning Models

ShadowBench: Exposing Lexical Anchoring and the Illusion of Forgetting in Large Language Models