Search papers, labs, and topics across Lattice.
Affiliation:
3
0
7
1
WnW reduces GPU memory usage to 20% of audio tokens without sacrificing accuracy, challenging the limitations of existing KV cache methods in long-form speech processing.
System Intelligence emerges as a game-changer, enabling LLM agents to collaborate effectively across complex tasks by organizing their interactions through dynamic graph structures.
Fake news detection gets a boost: RASR leverages retrieval-augmented semantic reasoning and domain-guided multimodal reasoning to significantly outperform existing methods, achieving up to 0.93% accuracy improvement.