Division-of-Thoughts reduces LLM API cost and latency by about 84% and 66% on seven reasoning benchmarks via subtask decomposition and small/large model routing, with accuracy near cloud-only baselines.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Division-of-Thoughts: Harnessing Hybrid Language Model Synergy for Efficient On-Device Agents
Division-of-Thoughts reduces LLM API cost and latency by about 84% and 66% on seven reasoning benchmarks via subtask decomposition and small/large model routing, with accuracy near cloud-only baselines.