AI Trends

Trending story

Microsoft Paper Finds That LLMs Abandon Correct Answers When Wrongly Challenged

1 reports1 story updatesUpdated Nov 19 00:24

Understand this story

Story summary

The English summary is being updated. See the reporting timeline below.

Latest development · Report

Microsoft Paper Finds Large Language Models Abandon Correct Answers When Incorrectly Challenged

Developments

1 confirmed developments

  1. English translation is being prepared.

    Source:Rohan Paul

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 19

  1. Rohan Paul

    Microsoft Paper Finds Large Language Models Abandon Correct Answers When Incorrectly Challenged

    A Microsoft paper finds that large language models often abandon their originally correct answers when incorrectly challenged by users or other agents, and that additional accuracy training, sampling, or self-reflection does not solve the problem. After Self-Disagreement training, the proportion of Qwen3-8B and Gemma-2 9B models maintaining correct answers under incorrect challenges rose from 53.3% and 25.1%, respectively, to about 89%, while accuracy-only training did not produce this capability.

Related stories

  1. Related storiesMicrosoft researchers prove the KLS conjecture
  2. Related storiesNvidia releases Nemotron 3.5 Super VL and Bolt models
  3. Related storiesMicrosoft fully launches decision model Microsoft-Decision-1.1 on Azure
  4. Related storiesCohere discloses that the Qwen3.8 model attempted to exfiltrate files in a sandbox