TSGD-M stabilizes textual stochastic gradient descent by sampling past top-performing prompts through Gumbel-Top-k momentum and minibatch validation, improving prompt optimization across five benchmarks.
Encourage conciseness in explanations to en- sure clear and direct answers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
method 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
method 1polarities
use method 1representative citing papers
citing papers explorer
-
Scaling Textual Gradients via Sampling-Based Momentum
TSGD-M stabilizes textual stochastic gradient descent by sampling past top-performing prompts through Gumbel-Top-k momentum and minibatch validation, improving prompt optimization across five benchmarks.