Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
No Free Lunch: Fundamental Limits of Learning Non-Hallucinating Generative Models
Authors: Changlong Wu, Ananth Grama, Wojciech Szpankowski
ICLR 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | In this paper, we develop a theoretical framework to analyze the learnability of nonhallucinating generative models from a learning-theoretic perspective. Our results reveal that non-hallucinating learning is statistically impossible when relying solely on the training dataset, even for a hypothesis class of size two and when the entire training set is truthful. Our findings are primarily conceptual, they represent a first step towards a principled approach to addressing hallucinations in learning generative models. |
| Researcher Affiliation | Academia | Changlong Wu1, Ananth Grama1 & Wojciech Szpankowski1,2 1CSo I, Purdue University 2Jagiellonian University EMAIL EMAIL |
| Pseudocode | No | The paper describes learning rules in mathematical equations (e.g., equations 7 and 8) but does not present any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not contain any explicit statements about releasing code, nor does it provide links to source code repositories. |
| Open Datasets | No | The paper focuses on theoretical analysis of generative models and the learnability of non-hallucinating models. It does not mention or use any specific datasets for empirical evaluation, thus no information about public availability of datasets is provided. |
| Dataset Splits | No | The paper is theoretical and does not involve empirical experiments with datasets, therefore, no dataset splits are discussed or provided. |
| Hardware Specification | No | The paper is theoretical and does not describe any experimental setup or results that would require specific hardware. No hardware specifications are mentioned. |
| Software Dependencies | No | The paper is theoretical and does not describe any experimental setup that would require specific software dependencies. No software names with version numbers are mentioned. |
| Experiment Setup | No | The paper presents a theoretical framework and mathematical proofs. It does not include any experimental results or detailed setup configurations such as hyperparameters or training schedules. |