Search papers, labs, and topics across Lattice.
3
0
4
0
The findings show that even a direction orthogonal to a concept at one layer can contribute to the concept's downstream amplification, and demonstrate that more fine-grained category residuals should also be considered beyond shared general harmfulness representation to fully understand LLM safety.
Post-editing attacks can destabilize watermark detection, but L-VQVAE's localized approach ensures reliable provenance in generative models.
Channel-specific historical data retrieval boosts multivariate time series forecasting accuracy, finally accounting for the fact that different variables have different rhythms.
