Qi Tian

Emotion recognition in visual intelligence is heavily skewed towards linguistic cues, revealing a critical gap in purely visual affect recognition capabilities.

Visual perturbations can significantly alter predictions in world models, but ACPC offers a quantifiable way to diagnose and mitigate these effects.

PairCoder boosts artifact verifiability by up to 3.9 times compared to traditional single-pass inference, revealing the power of collaborative programming in AI-generated outputs.