ContextAdapt evaluates LLMs on value alignment across medicine, law, finance, and national security
The paper introduces ContextAdapt, an evaluation framework to test whether large language models appropriately apply values like honesty, autonomy, and confidentiality across professional domains while remaining consistent when context does not change the relevant norm. Researchers evaluated 12 LLMs on recommended actions and justifications using scenarios based on primary-source professional…
Key points
- ContextAdapt evaluates LLMs on honesty, autonomy, and confidentiality across medicine, law, finance, and national security
- Models achieved 95.6% mean appropriateness but correct justification ranged from 25.6% to 76.9%
- Varying stakes caused localized failures where models changed responses despite unchanged obligations
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Anthropic's Claude begins 'dreaming' to reorganize memory in tests · 1 src
- John Langford quits Microsoft, declares AI Frontiers over · 1 src
- Anthropic releases study on Claude usage in scientific research · 3 src
- MIT and BASF release CheMeleon for low-data molecular prediction · 2 src
- Anthropic tests Claude agents in EVE Online trading and sabotage detection · 1 src
Comments
via GitHub Discussions