Pardis Sadat Zahraei

PardisSzah

·

pardissz

AI & ML interests

None yet

Recent Activity

authored a paper 20 days ago

Emergent Unfaithfulness: How Alignment Training Causes Language Models to Silently Override Task Faithfulness

authored a paper 20 days ago

AgentAbstain: Do LLM Agents Know When Not to Act?

authored a paper 20 days ago

I Am Aligned, But With Whom? MENA Values Benchmark for Evaluating Cultural Alignment and Multilingual Bias in LLMs

View all activity

Organizations

PardisSzah 's datasets 8

PardisSzah/fairstress

Viewer • Updated 26 days ago • 13M • 468

PardisSzah/fairstress-core

Viewer • Updated 26 days ago • 48k • 42

PardisSzah/FaithConflict

Viewer • Updated Oct 3 • 2.82k • 109

PardisSzah/ConvinceQA

Viewer • Updated Jul 1 • 6.94k • 42 • 1

PardisSzah/TWC

Viewer • Updated May 18, 2025 • 3.95k • 19 • 2

PardisSzah/DiseaseDecipher

Viewer • Updated Dec 23, 2024 • 12.7k • 45 • 2

PardisSzah/BiasMD

Viewer • Updated Sep 27, 2024 • 6.01k • 32 • 5

PardisSzah/DiseaseMatcher

Viewer • Updated Sep 22, 2024 • 32k • 34