Producing Structured Outputs from LLMs with Constrained Sampling
Summary
Discusses methods for generating structured outputs from large language models using constrained sampling techniques.
Similar Articles
Recipes for Steering and Scaling LLMs via Sampling
The paper presents a theoretical framework for steering and scaling large language models via sampling algorithms, such as Sequential Monte Carlo and Replica Exchange, to improve generation quality without external supervision.
Large Language Models for Token-Efficient and Semantic-Preserving Opinion Summarization
This paper presents a framework for opinion summarization using LLMs that combines multidimensional classification and stratified sampling to reduce token usage while preserving semantic diversity and balance across viewpoints.
Sequential statistical inference for Large Language Models: Representation, validity, and monitoring
This paper argues for a sequential inference framework to enhance LLM trustworthiness by modeling interactions as dependent stochastic processes, ensuring validity under repeated use, and enabling online monitoring for behavioral shifts.
Generating Robust Portfolios of Optimization Models using Large Language Models
Proposes a method to generate portfolios of optimization models using LLMs, with theoretical guarantees and empirical validation.
Synthetic Consumer Insight Generation with Large Language Models
This research examines whether LLMs can generate synthetic consumer data for projective techniques, comparing human and LLM responses on city tourism perceptions and finding substantial overlap but differences in style and diversity.