Xingwei Tan

XingweiT

·

Xingwei-Tan

AI & ML interests

None yet

Recent Activity

authored a paper about 15 hours ago

Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability

upvoted a paper 5 days ago

Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability

upvoted a paper 9 days ago

Synthetic Pre-pretraining Survives Scale, but Not as a Grammatical Prior

View all activity

Organizations

None yet

XingweiT 's papers 11

arxiv:2610.03509

arxiv:2606.08362

arxiv:2604.27251

arxiv:2604.16027

arxiv:2602.08658

arxiv:2602.03709

arxiv:2601.05882

arxiv:2510.03528

arxiv:2509.06652

arxiv:2508.19827

arxiv:2505.20415