Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts Paper • 2610.06147 • Published 12 days ago
Train Overcomplete, Deploy Compact: Scaling Recovery Capacity for Structured LLM Pruning Paper • 2609.06974 • Published Sep 7
Re-calibrated Contrastive Loss for Transformation-Aware Prompt Conditioning in Vision-Language Models Paper • 2609.06967 • Published Sep 7
meta-llama/Llama-4-Scout-17B-16E-Instruct Image-Text-to-Text • 109B • Updated May 22, 2025 • 163k • • 1.38k