Search papers, labs, and topics across Lattice.
HKUST (GZ)
4
0
6
The current RL checkpoint outperforms larger LLMs in redesigning training environments, revealing that iterative learning enhances diagnostic capabilities.
CoT fine-tuning can slash long-range recall by over 57% in hybrid LLMs, but a simple parameter restoration method can reverse this trend without additional training.
Loop-aware planning combined with descriptor-aided localization cuts exploration time by 15% and travel distance by 14% in challenging environments.
Ditch the heavy embedding sync: this federated recommendation method slashes communication costs by transmitting lightweight cluster labels instead of full item embeddings, without sacrificing accuracy.
