{
  "schema_version": "1.1",
  "id": "archive:https://arxiv.org/abs/2611.08433v1",
  "slug": "2611-08433v1-0oq3wi7",
  "url": "https://feed7.dev/p/2611-08433v1-0oq3wi7",
  "title": "MedGraph: Maintaining Patient State with Graph Memory for Long-Term Medical Dialogue",
  "why_included": "MedGraph organizes longitudinal dialogue as dated, person-attributed chains and retrieves them at the time a question asks about. The paper reports higher accuracy than five baselines on a medical dialogue benchmark.",
  "summary": "MedGraph stores dated, person-attributed medical facts and links medications, diagnoses, and measurements into chains. Retrieval ranks patient chains, selects the state relevant to the question's time, and adds original dialogue text; a patient profile supplies recorded safety facts.\nFor builders of agents with persistent memory, the design highlights three failure cases to guard against: outdated information, facts about the wrong person, and missing safety details. The paper reports **10.5 and 10.4 points** above the strongest baseline on MedMemoryBench with GPT and Qwen backbones, respectively.\nThese results are specific to longitudinal medical dialogue. The material does not establish performance in other domains or provide details about deployment with real patient data.",
  "practical_implication": "",
  "agent_context": "MedGraph stores dated, person-attributed medical facts and links medications, diagnoses, and measurements into chains. Retrieval ranks patient chains, selects the state relevant to the question's time, and adds original dialogue text; a patient profile supplies recorded safety facts.\nFor builders of agents with persistent memory, the design highlights three failure cases to guard against: outdated information, facts about the wrong person, and missing safety details. The paper reports **10.5 and 10.4 points** above the strongest baseline on MedMemoryBench with GPT and Qwen backbones, respectively.\nThese results are specific to longitudinal medical dialogue. The material does not establish performance in other domains or provide details about deployment with real patient data.",
  "source": {
    "name": "arXiv",
    "url": "https://arxiv.org/abs/2611.08433v1",
    "published_at": "2026-11-05T18:27:20.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Paper",
  "layer": "context",
  "domains": [
    "research",
    "data"
  ],
  "topics": [
    "agent-memory",
    "retrieval"
  ],
  "verification": {
    "status": "needs_review",
    "label": "Needs Review",
    "method": "unverified",
    "verified_at": null
  },
  "uncertainty": [],
  "connected_context": {
    "meaning": "MedGraph adds a domain-tested design for persistent memory that preserves dates, people, relationships, and relevant dialogue when reconstructing state. It sharpens general source-linked memory patterns with explicit safeguards against stale, misattributed, or incomplete facts, though its reported gains apply to longitudinal medical dialogue.",
    "corpus_size": 1020,
    "generated_at": "2026-11-06T18:41:22.072Z",
    "connections": [
      {
        "title": "nashsu/llm_wiki",
        "source_name": "GitHub",
        "source_url": "https://github.com/nashsu/llm_wiki",
        "feed7_url": "https://feed7.dev/p/llm-wiki-1dqmknq",
        "reason": "Both use structured external memory for agent retrieval; MedGraph adds time- and person-aware state chains, while LLM Wiki emphasizes source-linked compiled knowledge and review."
      },
      {
        "title": "Citation Needed: Provenance for LLM-Built Knowledge Graphs — Daniel Chalef, Zep AI",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=H7puB0RwJMM",
        "feed7_url": "https://feed7.dev/p/citation-needed-provenance-for-llm-built-knowledge-graphs-daniel-chalef-1iob5t8",
        "reason": "MedGraph’s dated, attributed facts and dialogue links provide a concrete structure for provenance concerns raised by the talk about merged or changing facts in knowledge graphs."
      },
      {
        "title": "How V7 gives AI agents institutional memory",
        "source_name": "OpenAI",
        "source_url": "https://openai.com/index/v7",
        "feed7_url": "https://feed7.dev/p/v7-1qw5ina",
        "reason": "Both present source-linked context as agent memory; MedGraph makes temporal and person attribution explicit, while V7’s summary leaves freshness and contradiction handling as open trust requirements."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-11-05T18:27:20.000Z",
  "modified_at": "2026-11-05T18:27:20.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/2611-08433v1-0oq3wi7",
    "json": "https://feed7.dev/p/2611-08433v1-0oq3wi7.json",
    "markdown": "https://feed7.dev/p/2611-08433v1-0oq3wi7.md"
  }
}