arxiv:2606.26080
🔄 In a Training Loop
Changdae Oh
changdae
AI & ML interests
Distribution Shift; Uncertainty Quantification; Reward Modeling
Recent Activity
authored a paper 1 day ago
Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring authored a paper 1 day ago
Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents upvoted a collection 1 day ago
WTF GENIUS PAPERSOrganizations
datasets 7
changdae/tau2-uq-artifacts
Viewer • Updated • 2.26M • 128
changdae/vittle-pope-visual-perturbed
Updated • 34
changdae/vittle-llavabench-coco-textual-perturbed
Viewer • Updated • 30 • 66
changdae/vittle-llavabench-coco-joint-perturbed
Viewer • Updated • 90 • 43
changdae/vittle-llavabench-coco-visual-perturbed
Viewer • Updated • 270 • 34
changdae/llavabench-shift-natural-v1
Viewer • Updated • 1.47k • 128
changdae/llavabench-shift-synthetic-v1
Updated • 206