Jailbreaking as a Reward Misspecification Problem
Paper
•
2406.14393
•
Published
•
12
•
2
Xie
Zhihui
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
10 days ago
OS-Genesis: Automating GUI Agent Trajectory Construction via Reverse
Task Synthesis
upvoted
a
paper
19 days ago
Diving into Self-Evolving Training for Multimodal Reasoning
updated
a Space
19 days ago
MMInstruction/VL-RewardBench
Organizations
Zhihui's activity
fix preprocessing errors
#1 opened 19 days ago
by
Zhihui
Remove redundant fields
#4 opened about 1 month ago
by
Zhihui
Cleanup parquet data
#3 opened about 1 month ago
by
Zhihui
Upload parquet version of the dataset
1
#2 opened 3 months ago
by
davanstrien
commented
2
papers
7 months ago
'QWenTokenizer' object has no attribute 'IMAGE_ST'
4
#1 opened about 1 year ago
by
Neman
'QWenTokenizer' object has no attribute 'IMAGE_ST'
4
#1 opened about 1 year ago
by
Neman