Cosine Misleads: Auxiliary Losses Reshape Vision Language Models, Not Their Latents
Paper • 2606.05753 • Published • 1
XiuYu Zhang
xiuyuz
AI & ML interests
None yet
Recent Activity
upvoted a paper about 11 hours ago
Cosine Misleads: Auxiliary Losses Reshape Vision Language Models, Not Their Latents submitted a paper about 11 hours ago
Cosine Misleads: Auxiliary Losses Reshape Vision Language Models, Not Their Latents upvoted a paper about 11 hours ago
Self-Evaluation Is Already There: Eliciting Latent Judge Calibration in Base LLMs with Minimal DataOrganizations
None yet
upvoted 2
papers about 11 hours ago