Poster Thu, Sep 10, 2026 • 1:30 AM – 3:30 AM PDT ExHall # 394

DIVA: Instruction-Aware Vision Token Pruning via Dual-Probe Attention Discrepancy

Hyunwoo Kim ⋅ Kisu Lee ⋅ Yuna Shin ⋅ Ha Young Kim

Keywords: Vision-Language Models & Foundation Models