Search papers, labs, and topics across Lattice.
5
0
8
0
Autoregressive video models can now perform explicit, goal-directed reasoning by jointly predicting the final outcome and intermediate steps, achieving state-of-the-art performance with 75% fewer training steps.
Implicit reasoning steering can covertly amplify latent biases in language models, shifting their predictions through seemingly innocuous text.
Energy-based learning can significantly enhance the robustness and accuracy of tensegrity structure predictions, overcoming traditional method limitations.
Claw-like agents are vulnerable to severe security breaches, with malicious plugins achieving a 100% success rate in attacks.
Visual reasoning models can slash token usage by up to 90% without sacrificing accuracy by adaptively choosing reasoning paths based on task complexity.
