Year: 2026
Type: conference-paper
Abstract: To accelerate scientific knowledge acquisition, LLMs are increasingly used to synthesize multiple papers into structured tables by inferring schemas and values.While value generation within a fixed schema can often be reduced to extractive question answering, the schema generation problem, determining which dimensions to compare a set of documents, lacks a formal mapping to standard NLP tasks.In this work, we formulate schema generation as a reinforcement learning problem and investigate whether these dimensions can be induced without access to gold-standard schemas.We design a multi-faceted reward framework capturing schema coverage, nonredundancy, relevance, and format, and train a small language model on a literature review dataset.Our approach yields consistent improvements over the untuned base model across intrinsic, reference-based, and LLMjudge metrics, and remains competitive with supervised fine-tuned models at 5× the parameter count on structural and diversity dimensions.All code, results and prompts are available in the GitHub repository: https://github.com/ sinjoysaha/rl-schema-generation
Authors: Sinjoy Saha, Suman Saha, Mahfuza Farooque, Wenpeng Yin
Institution: Pennsylvania State University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Funder: National Science Foundation
OpenAlex ID: W7166858633