Search papers, labs, and topics across Lattice.
5
1
9
9
vToken slashes KV memory retention by over 70%, dramatically boosting throughput and concurrency in large language model serving.
SHAPER enables embodied agents to self-evolve their skills and context without retraining, unlocking new possibilities for adaptation in fixed-interface scenarios.
Spurious correlations in foundation models can be effectively disentangled using a dual-branch approach, achieving superior bias mitigation with minimal parameter adjustments.
Off-policy degree in RLVR updates can drastically change which tokens drive learning, leading to a new adaptive method that outperforms traditional baselines.
LLM agents can learn to explore novel states and generalize to new tasks with a hybrid on- and off-policy RL framework that leverages memory.
