Generated Context versus Governed State: Functional Conditions for Accountable Longitudinal Clinical Reasoning

Augusto Bernardo Pissarra; Victor Lorena de Farias Souza

cs.AIAug 14, 2026

Generated Context versus Governed State: Functional Conditions for Accountable Longitudinal Clinical Reasoning

Authors: Augusto Bernardo Pissarra, Victor Lorena de Farias Souza

Organizations: MyndwareMed

Abstract

Large language models (LLMs) have become the dominant interface of clinical artificial intelligence, yet the interface they expose (text in, text out, one context window at a time) maintains no explicit, persistent, governed representation of what is currently true about a patient. This paper argues that longitudinal clinical reasoning is a state-estimation problem under partial observability, and that the axis on which clinical AI succeeds or fails is not the fluency of the model reading the record but the governance of the patient state it reasons over. We distinguish generated context from governed state; separate five objects that clinical AI habitually conflates (true state, observations, evidence, belief, and simulated state); define a tiered governance standard against which any clinical AI system can be audited; and show that an operational definition of accountability decomposes into four information requirements: an immutable evidence ledger with awareness-time versioning, a belief state distinct from accumulated evidence, an observation-process model, and claim-level causal typing. We are explicit that this decomposition is analytic rather than a necessity theorem, and that its value is conceptual hygiene: it converts "accountable clinical AI" from a slogan into an audit instrument. A six-level maturity framework separates what a system makes governable from what it can compute, locating current LLM-centric practice at high capability but low maturity. The paper is fully self-contained: the four research questions the framework poses are stated in the introduction, and the conclusion records what the paper establishes toward each; future work develops the buildable core of the architecture and the research program toward full Clinical World Models. No empirical result is claimed here.

Figures & tables

Figure 1: Generated context versus governed state. Left: knowledge of the patient is reconstituted into a transient buffer on every call and vanishes with it. Right: one canonical, versioned belief state — fed by an immutable evidence ledger — is the object every component reads from and writes through, and the object an auditor queries.

Tier	Requirements
Core	identity $\cdot$ persistence $\cdot$ explicit semantics $\cdot$ provenance $\cdot$ update governance $\cdot$ historical query (served by deterministic replay)
Safety-critical	core $+$ explicit uncertainty typing $\cdot$ deterministic constraint validation $\cdot$ correction propagation
Predictive	safety-critical $+$ outcome calibration $\cdot$ model-version binding $\cdot$ monitored validity under distribution shift

Table 1: The tiers of Definition 1 (Governed Clinical State), normalized. Each tier inherits the tier below it; a system is audited at the tier matching what its state is used for.

Architectural class	Persistent	Governed	Replay	Belief $\neq$ evid.	Quar. sim.
Electronic health record (EHR) / document store [ 19 ]	yes	partial	no	no	no
RAG / long-context LLM [ 4 ]	no	no	no	no	no
LLM agent $+$ memory [ 20 , 21 ]	partial	no	no	no	no
Knowledge graph (typical) [ 5 ]	yes	partial	no	no	no
Learned world model [ 11 , 22 , 23 ]	partial	no	no	no	partial
Governed clinical state (this paper)	yes	yes	yes	yes	yes

Table 2: Architectural classes, not products, against the governance properties of Definition 1 (Governed Clinical State). The five columns compress the audit for readability rather than replacing it: Persistent summarizes identity and persistence; Governed summarizes explicit semantics, update governance, provenance, deterministic constraints, and calibration; Replay summarizes historical query and rollback/correction; the last two columns audit the object separations of Section 4 (belief distinct from evidence; quarantined simulation), which the tier requirements presuppose. Entries indicate properties guaranteed by the architectural class as such, not properties a particular implementation could add; “partial” marks properties implementations may add without a class-level guarantee; the final row states what this paper requires, not what any deployed system has demonstrated — the audit questions, not the grades, are the contribution.

Figure 2: The five objects and the epistemic boundary. The true state $x_{t}$ sits above a boundary no system crosses; observations $o_{t}$ , evidence units $e_{i}$ , and belief $b_{\tau}(x_{t})$ are successive constructions below it, and simulated state $\tilde{x}^{\,\mathrm{branch}}_{t}$ lives in a quarantined branch whose values never re-enter evidence or belief. The pathologies of LLM-centric practice are conflations of adjacent stages — most commonly, treating evidence as if it were belief.

Component	Generated context (RAG / long-context)	Governed state
Renal function	Treats a 14-month-old creatinine as the patient’s current renal status — fluent, confident, stale	Belief over renal function widened by 14 months without measurement: “unknown; last known value $x$ ; a point-of-care creatinine would resolve it”
Metformin	Whichever fragment ranks higher wins; the conflict itself is invisible	Contradicted-open: both evidence chains preserved, belief explicitly uncertain, conflict flagged for human closure
Prior reaction	Renders absence as “no known allergies”	Typed unknown/unavailable, not negative: prior contrast records exist at a source unavailable to this system
The gaps themselves	No mechanism: retrieval returns documents, not the absence of documents	First-class output: the typed gaps and the measurements that would close them

Table 3: One clinical question, component by component. The failure on the left is not that the language model reads badly — it reads perfectly. The failure is that four of the five objects of Section 4 do not exist anywhere in that architecture, so no amount of reading can consult them.

Figure 3: One patient, six years, one query. Top: the timeline, with a documentation lag (2022), an availability exit (2023), and an informative measurement gap (2023–24). Bottom: the same record read at the same moment by the two architectures of Figure 1 . The governed answer is not produced by a better reader but by objects the generated-context architecture does not possess. The case is an illustrative construction, not an executed comparison: the claim is about mechanisms each architecture can and cannot express, not measured performance.

Figure 4: The six-level framework: increasing degrees of explicit, persistent, governable clinical structure. An architectural maturity framework, not a claimed law of intelligence.

AI Accountability

AI accountability concerns how responsibility for AI decisions and harms is assigned, evidenced, challenged, and remedied.

See all 157 papers

Can LLMs Reason Over Long Horizons? An Empirical Evaluation of Context Strategies for Longitudinal Clinical Reasoning
Sep 30, 2026Taye Akinrele, Noorbakhsh Amiri Golilarz, Subash Neupane +2Long-Context Language Model Evaluation Clinical Reasoning
Department of Computer Science, The University of Alabama
A rubric landscape for evaluating clinical reasoning in large language models: what exists, what is missing, and what needs to be combined
Oct 1, 2026Zhangshu Joshua Jiang, Zina Ibrahim, James T. TeoFaithfulness of Language Model Explanations Temporal Reasoning in Language Models
DRIVE-Health CDT Department of Biostatistics and Health Informatics Institute of Psychiatry, Psychology and Neuroscience King’s College London London, UK · Cleveland Clinic London London, UK
LAAF: A Layered Accountability Architecture Framework for LLM Applications
Aug 27, 2026Prachi Chaturvedi, Shahnawaz Ahmad, Ehsan Nowroozi +5AI Accountability Algorithmic Accountability
School of Computer Science Engineering and Technology, Bennett University, Greater Noida, India · Centre for Sustainable Cyber Security (CS2), University of Greenwich, United Kingdom · School of Electrical Engineering, Computing & Mathematical Sciences, Curtin University, Australia +1

Sep 30, 2026J/K move · Enter open · S save

Can LLMs Reason Over Long Horizons? An Empirical Evaluation of Context Strategies for Longitudinal Clinical Reasoning

Taye Akinrele, Noorbakhsh Amiri Golilarz, Subash Neupane, Sudip Mittal+1

Department of Computer Science, The University of Alabama

Long-Context Language Model Evaluation Clinical Reasoning LLM Context Management Medical LLMs

Open paper AskPDF arXiv

Longitudinal clinical reasoning requires large language models (LLMs) to identify and integrate relevant evidence distributed across extended patient histories. Although long-context models can process increasingly large amounts of information, providing more history does not necessarily make relevant evidence more accessible or improve reasoning. We compare five context strategies (Full, Recent, Episodic, Semantic, and Hybrid) on MedLoCoMo across four open-weight LLMs, examining answer correctness, robustness to query-evidence distance, and abstention on questions with unsupported premises. Episodic and Hybrid generally achieve the strongest overall accuracy, while Recent Context degrades most as supporting evidence becomes more distant; Episodic and Hybrid maintain the highest accuracy at long distances. Analysis of adversarial questions further shows that strong performance on answerable questions does not necessarily translate to successful abstention when the available history does not support the requested conclusion. These findings show that reliable longitudinal reasoning depends not only on how much history an LLM can access, but critically on how relevant evidence is selected and presented for reasoning.

Abstract

AI Accountability

Explore similar work

Can LLMs Reason Over Long Horizons? An Empirical Evaluation of Context Strategies for Longitudinal Clinical Reasoning

A rubric landscape for evaluating clinical reasoning in large language models: what exists, what is missing, and what needs to be combined

LAAF: A Layered Accountability Architecture Framework for LLM Applications