EnvScaler synthesizes 191 environments and 7K scenarios to improve LLM agents' multi-tool interaction performance via SFT and RL on Qwen3 models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis
EnvScaler synthesizes 191 environments and 7K scenarios to improve LLM agents' multi-tool interaction performance via SFT and RL on Qwen3 models.