# Reference-Aware Language Models Source: [Reference-Aware Language Models]() Collection index: [llms.txt]() Data updated: 2026-01-01T12:39:45.087Z Authors: Zichao Yang, Phil Blunsom, Chris Dyer, Wang Ling Published: 2016-11-05T10:55:37.000Z ## Abstract We propose a general class of language models that treat reference as an explicit stochastic latent variable. This architecture allows models to create mentions of entities and their attributes by accessing external databases (required by, e.g., dialogue generation and recipe generation) and internal state (required by, e.g. language models which are aware of coreference). This facilitates the incorporation of information that can be accessed in predictable locations in databases or discourse context, even when the targets of the reference may be rare words. Experiments on three tasks shows our model variants based on deterministic attention. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Recipe Generation on Allrecipes (val)](): Recipe Generation; Allrecipes (val); PPL (All): 5.02. - [Recipe Generation on Allrecipes (test)](): Recipe Generation; Allrecipes (test); PPL (All): 4.97. - [Language Modeling on Xinhua News (val)](): Language Modeling; Xinhua News (val); PPL (All): 30.43. - [Language Modeling on Xinhua News (test)](): Language Modeling; Xinhua News (test); PPL (All): 30.42. ## Original sources and implementation [Original paper]() [PDF]()