Grounding Without Corrective Control: Truth-Tracking Profiles for Large Language Models
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 11 of its 53 references that are in the map. First posted on arXiv 14 Aug 2026.
Builds on
11 of 53 references.
References · 11
- 2023Evaluating Object Hallucination in Large Vision-Language Modelscited by 1,729
- 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
- 2021Retrieval Augmentation Reduces Hallucination in Conversationcited by 692
- 2022Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Taskcited by 431
- 2023The Vector Grounding Problemcited by 39
- 2026How RLHF Amplifies Sycophancycited by 23
- 2023Do Language Models' Words Refer?cited by 17
- 2024Are Language Models More Like Libraries or Like Librarians? Bibliotechnism, the Novel Reference Problem, and the Attitudes of LLMscited by 11
- 2024Does Thought Require Sensory Grounding? From Pure Thinkers to Large Language Modelscited by 7
- 2024Large language models and linguistic intentionalitycited by 6
- 2023Meaning and understanding in large language modelscited by 5
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026The Confidence Manifold: Geometric Structure of Correctness Representations in Language Models
- 2026Linear Separability of Activation Representations after Supervised Fine-Tuning on Incorrect Responses: A Study of Synthetic Dishonesty in Large Language Models
- 2025Depth-Wise Activation Steering for Honest Language Models
- 2026Locating Answer-Correctness Signals in Frozen Large Language Models
- 2026Confidence Geometry Reveals Trace-Level Correctness in Large Language Model Reasoning
- 2026The Knowing-Saying Gap: When Probes See Errors that Confidence Misses
- 2026The Attribution Blind Spot: Detecting When Language Models Rely on Memory Rather Than Retrieved Context
- 2025Internal Reasoning vs. External Control: A Thermodynamic Analysis of Sycophancy in Large Language Models