Grounding Without Corrective Control: Truth-Tracking Profiles for Large Language Models

Abstract:Recent work suggests that some large language model representations have content or reference. Grounding can secure either without supplying live routes for correction. This paper asks what follows from that gap. An output is answerable when discrepancies can affect what a target- and task-specific arrangement produces, accepts, or withdraws. The arrangement has corrective control only when live, sufficiently independent routes can detect and repair fresh discrepancies. A route profile records which routes constrain the arrangement and how they are related. Those profiles support analysis of truth-tracking: patterned support for representational success. Language models are the pressure case; text-only arrangements provide a task-relative limiting case. Text-trained models inherit patterns of testimony, coherence, and prior correction. Where target-sensitive correction survives training, these can supply derivative answerability (inherited constraint); live answerability is the relation supplied by a current route for fresh discrepancies. Fluent failures should follow when a task requires independently informative access to the facts. Self-consistency, retrieval, tools, code execution, multimodal input, and feedback should help selectively. Route-by-task interactions test the distinctions. The decomposition's empirical burden is to predict held-out route--task combinations or improve intervention choice without conceptual refitting. Surface improvement and truth-tracking improvement can come apart.

24 pages, 1 figure, 1 table. A six-page methodological supplement, reproducible R script, and constructed data are included as ancillary files

Artificial Intelligence (cs.AI); Computation and Language (cs.CL)

arXiv:2608.14252 [cs.AI]

(or arXiv:2608.14252v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2608.14252 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

gelman_fake_data_calibration.csv

gelman_fake_data_metadata.txt

gelman_fake_data_paired_comparisons.csv

gelman_fake_data_replications.csv

gelman_fake_data_summary.csv

gelman_fake_data_workflow.R

online-resource-1.pdf

(3 additional files not shown) You must enabled JavaScript to view entire file list.

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.