No items found.

Position: We Need A Unified Definition of Hallucination (It’s The World Model, Stupid!)

Emmy Liu, Varun Gangal, Chelsea Zou, Michael Yu, Xiaoqi Huang, Alex Chang, Zhuofu Tao, Karan Singh, Sachin Kumar, Steven Y. Feng

Emmy Liu, Varun Gangal, Chelsea Zou, Michael Yu, Xiaoqi Huang, Alex Chang, Zhuofu Tao, Karan Singh, Sachin Kumar, Steven Y. Feng

“Meighen Island”, Canadian Arctic
03

Abstract

Despite numerous attempts at mitigation since the inception of language models, hallucinations remain a persistent problem even in today’s frontier LLMs. Why is this? We review existing definitions of hallucination and fold them into a single, unified definition wherein prior definitions are subsumed. This position paper argues that hallucination can be unified by defining it as simply inaccurate (internal) world modeling, in a form where it is observable to the user. For example, stating a fact which contradicts a knowledge base OR producing a summary which contradicts the source. By varying the reference world model and conflict policy, our framework unifies prior definitions. We argue that this unified view is useful because it forces evaluations to clarify their assumed reference “world”, distinguishes true hallucinations from planning or reward errors, and provides a common language for comparison across benchmarks and discussion of mitigation strategies. Building on this definition, we also connect our framework to HalluWorld (Liu et al., 2026), a complementary benchmark that instantiates fully specified reference world models for stress-testing model hallucinations.

03

Runs

Cite this
TextBibTeXCopy
Emmy Liu, Varun Gangal, Chelsea Zou, Michael Yu, Xiaoqi Huang, Alex Chang, Zhuofu Tao, Karan Singh, Sachin Kumar, Steven Y. Feng. Position: We Need A Unified Definition of Hallucination (It’s The World Model, Stupid!). In Proceedings of the 43rd International Conference on Machine Learning, volume 306 of Proceedings of Machine Learning Research, pages 171298–171314, 2026.
@InProceedings{pmlr-v306-liu26gu, title = {Position: We Need A Unified Definition of Hallucination ({I}t’s The World Model, Stupid!)}, author = {Liu, Emmy and Gangal, Varun Prashant and Zou, Chelsea and Yu, Michael and Huang, Xiaoqi and Chang, Alex and Tao, Zhuofu and Singh, Karanpartap and Kumar, Sachin and Feng, Steven Y.}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {171298--171314}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/liu26gu/liu26gu.pdf}, url = {https://proceedings.mlr.press/v306/liu26gu.html}, abstract = {Despite numerous attempts at mitigation since the inception of language models, hallucinations remain a persistent problem even in today’s frontier LLMs. Why is this? We review existing definitions of hallucination and fold them into a single, unified definition wherein prior definitions are subsumed. This position paper argues that hallucination can be unified by defining it as simply inaccurate (internal) world modeling, in a form where it is observable to the user. For example, stating a fact which contradicts a knowledge base OR producing a summary which contradicts the source. By varying the reference world model and conflict policy, our framework unifies prior definitions. We argue that this unified view is useful because it forces evaluations to clarify their assumed reference “world”, distinguishes true hallucinations from planning or reward errors, and provides a common language for comparison across benchmarks and discussion of mitigation strategies. Building on this definition, we also connect our framework to HalluWorld (Liu et al., 2026), a complementary benchmark that instantiates fully specified reference world models for stress-testing model hallucinations.} }
ICML 2026
December 25, 2025
ICML 2026
Main
Position paper track
Poster
Agentic AI in the Wild: From Hallucinations to Reliable Autonomy
Poster
unified-hallucination-framework
More research
All research →
“Serene Expressions”, off the Louisiana coast“Serene Expressions”, off the Louisiana coast
NeurIPS 2026
meta

HalluWorld: A Controlled Benchmark for Hallucination via Reference World Models

May 19, 2026
NeurIPS 2026
Main
halluworld
NeurIPS 2026
“Alluvial Fan”, Taklimakan Desert, China“Alluvial Fan”, Taklimakan Desert, China
ICML 2026
meta

Benchmarking Reward Hack Detection in Code Environments via Contrastive Analysis

January 27, 2026
ICML 2026
Main
trace
ICML 2026
From the blog
All posts →
No items found.