Ming Li

The paper develops a quantitative framework to analyze and improve flow distillation in diffusion models, providing stability guarantees and suggesting non-uniform time scheduling to reduce approximation errors.

The paper develops a quantitative framework to analyze and improve flow distillation in diffusion models, providing stability guarantees and suggesting non-uniform time scheduling to reduce approximation errors.

The paper analyzes the Environment-Aware Speech and Sound Deepfake Detection Challenge and summarizes effective design choices from top-performing submissions.

The paper introduces Epi2Diff, a framework that maps Large Reasoning Models reasoning traces into cognitively grounded episode sequences for human item difficulty prediction, outperforming strong baselines.

This paper introduces the REAL-TSE Challenge, a satellite challenge on target speaker extraction from real conversational recordings, and describes its task definition, datasets, evaluation protocol, and submitted systems.

This paper proposes a neural audio watermarking method that embeds a message into the continuous latent representation of a codec-like speech autoencoder for improved codec robustness.

This paper proposes a dual-branch visual front-end for Audio-Visual Target Speaker Extraction (AVTSE) that reduces visual encoding computation while maintaining performance.

This paper introduces EvoSpeakerMem, a training-free layer for personalized spoken agents, and LoCoMo-ID, a spoken-memory benchmark.