Explore
AI Tools
New Tools
AI Agents
One9 Worker
LLMs
RAG & Vector DBs
Research
Assemble a stack
Founder Stacks
Compare
How We Rate
About
$
/
₹
⌘K
Sign up free
Login
Back to papers
June 29, 2026
cs.LG
cs.AI
stat.ML
Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models
Subramanyam Sahoo
,
Aman Chadha
,
Vinija Jain
,
Divya Chaudhary
Original Abstract
Read on arXiv
Download PDF
Categories
cs.LG, cs.AI, stat.ML