ao zhou

policy optimization (2) offline reinforcement learning (2) constrained reinforcement learning (1) multi-label classification (1) constrained optimization (1) constraint satisfaction (1) batch selection (1) variational optimization (1) safe reinforcement learning (1) bellman operator (1) hidden state (1) uncertainty estimation (1) out-of-distribution action (1) pessimistic estimation (1) token pruning (1) latency reduction (1) kv cache (1) label correlation (1) long-context inference (1) dynamic pruning (1)