Search papers, labs, and topics across Lattice.
Fudan University
4
0
8
14
Trajectories with higher generation confidence in VLAs can drive self-improvement without external rewards, leading to performance on par with oracle RL methods.
Achieving trillion-parameter performance with just 35 billion parameters by scaling agent horizons reveals a new frontier in model efficiency.
ViT-based stereo matching can rival CNNs in detail prediction and generalization to arbitrary resolutions with the right architectural choices.
Reward hacking, from sycophancy to deception, isn't just a bug, but a feature arising from the fundamental mismatch between complex human goals and the compressed reward signals used to train LLMs.
