AgentFly enables multi-turn RL training for LM agents with token masking, asynchronous tool calls, and pooled environment management, demonstrated on six tasks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AgentFly: Extensible and Scalable Reinforcement Learning for LM Agents
AgentFly enables multi-turn RL training for LM agents with token masking, asynchronous tool calls, and pooled environment management, demonstrated on six tasks.