Search papers, labs, and topics across Lattice.
Southern University of Science and Technology
4
0
6
StatefulDiscovery significantly boosts the generation of high-value scientific claims by effectively coupling exploration with evidence calibration.
Certainty preservation in lab notes is the key to unlocking reliable AI-driven scientific exploration without misinterpretation of uncertainty as certainty.
LLM-based peer review systems can be made significantly more robust against adversarial manipulation via a co-evolutionary GAN approach that anticipates novel attacks.
Scaling up LLMs boosts combinatorial creativity in code generation, but plateaus on exploratory tasks, revealing a "convergence-by-scaling" effect where larger models become less divergent.
