Track new results across pure and applied mathematics.
Search papers, labs, and topics across Lattice.
South China University of Technology
4
0
6
A single-rollout RLVR method achieves faster training and more stable gradients by using a trajectory's own future as a hindsight context, eliminating the need for group sampling and reference solutions.
Aligning inference with the forward statistical structure of diffusion models can dramatically boost generative fidelity, as evidenced by significant performance gains on CIFAR-10 and ImageNet.
Conditioning on informative local measurements transforms a memorization-prone estimator into a high-fidelity quantum state tomography tool, achieving 0.95 fidelity with fewer measurements.
Entangled quantum circuits can significantly hinder generalization, leading to worse performance than non-entangled circuits with the same number of parameters.