H. Wang, “Emotion-Conditioned Motion Sub-spaces with Flow Matching for Real-Time Audio-Driven Talking Heads”, AAAI, vol. 40, no. 21, pp. 17769–17777, Mar. 2026.