An automatic differentiation-style framework optimizes prompts across multi-component and cyclic LLM workflows, outperforming single-node textual-gradient baselines on several small benchmarks.
Training your sparse neural network better with any mask
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
LLM-AutoDiff: Auto-Differentiate Any LLM Workflow
An automatic differentiation-style framework optimizes prompts across multi-component and cyclic LLM workflows, outperforming single-node textual-gradient baselines on several small benchmarks.