mlsanityData to 14 Oct 2026

Detecting Clinical Discrepancies in Health Coaching Agents: A Dual-Stream Memory and Reconciliation Architecture

arXiv 2604.27045 ↗First posted 29 Apr 2026cs.LG cs.AI cs.CLcited by 0

On the map: LLM agents and tool use → Medical and scientific domain agents

Abstract (arXiv)

As Large Language Model (LLM) agents transition from single-session tools to persistent systems managing longitudinal healthcare journeys, their memory architectures face a critical challenge: reconciling two imperfect sources of truth. The patient's evolving self-report is current but prone to recall bias, while the Electronic Health Record (EHR) is medically validated but frequently stale. General-purpose agent memory systems optimize for coherence by overwriting older facts with the user's latest statement, a pattern that risks safety failures when applied to clinical data. We introduce a Dual-Stream Memory Architecture that strictly separates the patient narrative from the structured clinical record (FHIR), governed by a dedicated Reconciliation Engine that evaluates every extracted memory against the patient's FHIR profile and classifies discrepancies by type, severity, and the specific FHIR resources involved. We evaluate this architecture on 26 patients across 675 longitudinal wellness coaching sessions, using a hybrid dataset that interleaves real provider-patient transcripts with synthetic, FHIR-grounded clinical scenarios. In isolated testing, the engine detects 84.4% of designed clinical discrepancies with 86.7% safety-critical recall. By coupling extraction and reconciliation evaluation on the same data, we directly quantify a 13.6% error cascade, tracing the degradation to clinical details lost during memory extraction from unstructured conversation rather than to downstream classification errors. These findings establish that validating patient-reported memories against clinical records is both feasible and necessary for safe deployment of longitudinal health agents.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use. Builds on 15 of its 37 references that are in the map. First posted on arXiv 29 Apr 2026.

Builds on

15 of 37 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025FHIR-AgentBench: Benchmarking LLM Agents for Realistic Interoperable EHR Question Answering12 Sep
  2. 2025AI Agents for Conversational Patient Triage: Preliminary Simulation-Based Evaluation with Real-World EHR Data4 Jun

Other references · 13

  1. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  2. 2023MemGPT: Towards LLMs as Operating Systemscited by 1,389
  3. 2019Publicly Available Clinical BERT Embeddingscited by 1,002
  4. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 830
  5. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memorycited by 779
  6. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 707
  7. 2025Zep: A Temporal Knowledge Graph Architecture for Agent Memorycited by 400
  8. 2020Med-BERT: pre-trained contextualized embeddings on large-scale structured electronic health records for disease predictioncited by 373
  9. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactionscited by 254
  10. 2025MemoryBench: A Benchmark for Memory and Continual Learning in LLM Systemscited by 61
  11. 2026BeliefShift: Benchmarking Temporal Belief Consistency and Opinion Drift in LLM Agentscited by 12
  12. 2024PrecLLM: A Privacy-Preserving Framework for Efficient Clinical Annotation Extraction from Unstructured EHRs using Small-Scale LLMscited by 4
1 more
  1. 2025Data Quality Challenges in Retrieval-Augmented Generationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Personalized State-Transition-Aware Memory for Clinical Agents
  2. 2026ClinicalAgents: Multi-Agent Orchestration for Clinical Decision Making with Dual-Memory
  3. 2025DualAlign: Generating Clinically Grounded Synthetic Data
  4. 2026CliniCARE-Bench: Clinical Calibrated Audit of Medical Reasoning in EHR
  5. 2025Patient-Zero: Scaling Synthetic Patient Agents to Real-World Distributions without Real Patient Data
  6. 2026Generated Context versus Governed State: Functional Conditions for Accountable Longitudinal Clinical Reasoning
  7. 2025TIMER: Temporal Instruction Modeling and Evaluation for Longitudinal Clinical Records
  8. 2026TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots