Search papers, labs, and topics across Lattice.
College of Computing and Data Science, Nanyang Technological University
2
0
3
Automated creativity evaluation reveals that LLMs can achieve significant improvements in task fulfillment efficiency, transforming how we assess AI creativity across diverse applications.
Context sensitivity in LLMs evolves significantly across training stages, revealing that SFT biases models towards simpler contexts that can be both reinforced and reshaped later on.
