Semantic Drift and the Stability of Operator Control in Reasoning-Class Decision Support Systems
On the map: LLM reasoning → Reasoning benchmarks and evaluation
Abstract (arXiv)
The article investigates the fundamental problem of ensuring the stability of operator control and preserving goal-targeting in hybrid human-machine decision support systems (DSS) of a new generation. Based on a two-month continuous longitudinal experiment on the joint design of a monograph-format textual array, the latent phenomenon of semantic context drift in large language models of deep logical reasoning (Reasoning LLMs) is verified and described. A mathematical model of interaction in the human-machine interface is proposed, and an original metric is introduced - the operator control stability coefficient, which takes into account the non-linear contextual pressure of hidden reasoning chains. Within the paradigm of the cognitome theory, a critical point of control functions inversion is captured. Engineering recommendations are formulated for implementing dynamic relational arbitration loops based on a modified hierarchical similarity model.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 9 of its 23 references that are in the map. First posted on arXiv 8 Jul 2026.
Builds on
9 of 23 references.
References · 9
- 2023Challenges and Applications of Large Language Modelscited by 251
- 2025A Survey of Process Reward Models: From Outcome Signals to Process Supervisions for Large Language Modelscited by 37
- 2025Drift No More? Context Equilibria in Multi-Turn LLM Interactionscited by 21
- 2024FocusLLM: Precise Understanding of Long Context by Dynamic Condensingcited by 10
- 2026The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steeringcited by 5
- 2026Reasoning Shift: How Context Silently Shortens LLM Reasoningcited by 3
- 2026Process Reward Models Meet Planning: Generating Precise and Scalable Datasets for Step-Level Rewardscited by 1
- 2026Process Supervision via Verbal Critique Improves Reasoning in Large Language Modelscited by 1
- 2026Verifiable Counterfactual Supervision for Process Reward Modelscited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026The Missing Knowledge Layer in AI: A Framework for Stable Human-AI Reasoning
- 2025On the Reasoning Capacity of AI Models and How to Quantify It
- 2025Beyond "Hallucinations": A Framework for Stable Human-AI Reasoning
- 2026Semantic Invariance in Agentic AI
- 2025Logical Reasoning in Large Language Models: A Survey
- 2026Knowledge Vector of Logical Reasoning in Large Language Models
- 2025ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning
- 2023Studying and improving reasoning in humans and machines