Search papers, labs, and topics across Lattice.
Toyota Technological Institute at Chicago
3
0
4
INFUSER outperforms a frozen 32B model with just an 8B co-evolving generator, showcasing the power of adaptive question generation in self-evolution.
Transformers require a surprisingly high number of examples for effective chain-of-thought learning, challenging assumptions about their efficiency.
Stop forcing object-centric learning models into rigid feature alignment; cycle consistency works better when applied to the reconstructed visual scene.
