REVIEW 6 cited by
Learning How to Ask: Querying LMs with Mixtures of Soft Prompts
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Natural-language prompts have recently been used to coax pretrained language models into performing other AI tasks, using a fill-in-the-blank paradigm (Petroni et al., 2019) or a few-shot extrapolation paradigm (Brown et al., 2020). For example, language models retain factual knowledge from their training corpora that can be extracted by asking them to "fill in the blank" in a sentential prompt. However, where does this prompt come from? We explore the idea of learning prompts by gradient descent -- either fine-tuning prompts taken from previous work, or starting from random initialization. Our prompts consist of "soft words," i.e., continuous vectors that are not necessarily word type embeddings from the language model. Furthermore, for each task, we optimize a mixture of prompts, learning which prompts are most effective and how to ensemble them. Across multiple English LMs and tasks, our approach hugely outperforms previous methods, showing that the implicit factual knowledge in language models was previously underestimated. Moreover, this knowledge is cheap to elicit: random initialization is nearly as good as informed initialization.
Forward citations
Cited by 6 Pith papers
-
Training-Free Token-Level Steering for LLM Personalized Co-Writing
A token-level, training-free steering framework that improves LLM personalized co-writing by mixing the base model's posterior with a kernel-density estimate from a small user dataset.
-
DetPO: In-Context Learning with Multi-Modal LLMs for Few-Shot Object Detection
Detection Prompt Optimization (DetPO) improves few-shot object detection with black-box MLLMs by iteratively refining text prompts from TP/FP/FN errors on few-shot examples, gaining up to 9.7 mAP over prior black-box methods.
-
An Efficient Evolutionary Algorithm for Few-for-Many Optimization
SoM-EMOA, a (μ+1) evolution strategy that directly minimizes the sum-of-minimum coverage objective, outperforms existing many-objective solvers on a new R2-based benchmark suite for few-for-many optimization.
-
Stabilizing Black-Box Prompt Optimization with Textual Regularization and Signal Aggregation
TRAS adds success-based textual regularization and Monte Carlo signal aggregation to black-box prompt optimization, improving accuracy and reducing instruction loss when moving prompts across models.
-
PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation
PAID proposes Householder-based orthogonal weight updates for continual test-time adaptation, claiming that preserving pairwise angular structure of pretrained weights is a useful prior, but the math and validation fo...
-
Evolutionary Computation and Large Language Models: A Survey of Methods, Synergies, and Applications
A survey that maps bidirectional synergies between evolutionary computation and large language models and proposes a taxonomy plus research gaps.
Discussion (0). Continue with ORCID to comment.