Wei Guo

Forget complex model architectures for cross-domain recommendation: Taesar shows that cleverly transforming your data can unlock better performance with standard models.

PINNs can be made dramatically more robust to noisy data (up to 96.6% error reduction) by selectively pruning neurons that are overly sensitive to corrupted samples.

Recommender systems can bootstrap their performance without external data via a recursive self-improvement loop that generates, filters, and learns from its own plausible user interaction sequences.

Platform-centric digital services may be prioritizing engagement over user well-being, but LLMs and on-device intelligence now make truly user-centric agents a feasible alternative.