Search papers, labs, and topics across Lattice.
4
0
5
0
Adversarial attacks can be effectively mitigated in LVLMs by jointly optimizing visual and semantic supervision, leading to enhanced robustness across diverse tasks.
CRF-based training for auditory attention decoding boosts accuracy by up to 5.6% over traditional methods, revealing the power of temporal context in neural signal interpretation.
Frontier LLMs can be induced to generate biologically hazardous sequences, with attack success rates reaching up to 100%.
Not every missing modality needs to be repaired for optimal sentiment analysis, and SIEVE learns to make this decision dynamically at the sample level.
