{
  "abstract": "Language models are often asked to produce complete answers, pages, reports, dashboards, or interfaces in one generation. This gives the model too much responsibility: it must select sources, preserve state, choose layout, generate all text, maintain internal consistency, and express the result in a renderable form. Small models suffer most, but large models also produce malformed structures, unsupported claims, and destructive rewrites. SurfaceMind separates intelligence from rendering and authority. It compiles exact sources and governed state into a versioned Semantic Surface Graph. A model operates the graph through small, schema-bound SurfaceDelta patches. A deterministic validator checks base version, target existence, citation spans, text budgets, ordering, authority, and surface invariants before the renderer updates the view. The model never generates authoritative pixels and never owns truth. The architecture externalizes working memory into a visible surface. A compact local model can classify, extract, rank, repair, or patch one region instead of regenerating an entire answer. The system preserves exact source spans, evidence, versions, omissions, and receipts. SurfaceMind's central empirical claim—matched 4B tasks perform better through SurfaceDelta than raw full-answer chat—remains a named go/no-go experiment, not a completed result.",
  "author": "Hayden Lindley",
  "category": "Operational Surfaces",
  "claims": {
    "claims": [
      {
        "claim_id": "P07-C01",
        "evidence_refs": [
          "P07-S01"
        ],
        "limitations": [
          "Integrated mobile runtime and broad surface library remain incomplete."
        ],
        "public": true,
        "scope": "R0 repository foundation.",
        "statement": "SurfaceMind R0 defines a typed Semantic Surface Graph, deterministic rendering/verification contracts, and verified example surfaces.",
        "status": "IMPLEMENTED"
      },
      {
        "claim_id": "P07-C02",
        "evidence_refs": [
          "P07-S02",
          "P07-S03",
          "P07-S04"
        ],
        "limitations": [
          "Production validators require exact repository receipts."
        ],
        "public": true,
        "scope": "Max-Flex and GlyphCAS architecture.",
        "statement": "SurfaceDelta provides an inspectable optimistic-concurrency mutation unit with source-span validation before rendering.",
        "status": "SPECIFIED"
      },
      {
        "claim_id": "P07-C03",
        "evidence_refs": [
          "P07-S02",
          "P07-S03"
        ],
        "limitations": [
          "This is the go/no-go hypothesis and must not be stated as achieved."
        ],
        "public": true,
        "scope": "E1 matched experiment.",
        "statement": "A 4B model operating through SurfaceDelta outperforms matched raw full-answer generation on reliability, citation, latency, and usefulness.",
        "status": "AWAITING_LOCAL_RUN"
      }
    ],
    "disclosure_note": "Claims marked non-public in the authored ledger are omitted from this public artifact.",
    "paper": "p07-surfacemind-surfacedelta",
    "version": "0.1.0",
    "withheld_claim_count": 0
  },
  "date": "2026-07-27",
  "diagrams": {
    "diagrams": [
      {
        "brief": "Sources → EvidenceCapsules → Semantic Surface Graph → SurfaceDelta validator → deterministic renderer → receipts.",
        "format": "svg",
        "id": "P07-F01",
        "path": "/diagrams/p07-surfacemind-surfacedelta/P07-F01.svg",
        "status": "source-generated conceptual architecture",
        "title": "SurfaceMind stack"
      },
      {
        "brief": "Large tangled generation path contrasted with one small patch to an existing block, with token and validation surfaces.",
        "format": "svg",
        "id": "P07-F02",
        "path": "/diagrams/p07-surfacemind-surfacedelta/P07-F02.svg",
        "status": "source-generated conceptual architecture",
        "title": "Full answer versus delta"
      },
      {
        "brief": "Visual legend for verified, derived, speculative, disputed, stale, and breach states.",
        "format": "svg",
        "id": "P07-F03",
        "path": "/diagrams/p07-surfacemind-surfacedelta/P07-F03.svg",
        "status": "source-generated conceptual architecture",
        "title": "Truth-state rendering"
      },
      {
        "brief": "Same 4B/source/task split into raw prose and SurfaceDelta conditions, converging on common metrics.",
        "format": "svg",
        "id": "P07-F04",
        "path": "/diagrams/p07-surfacemind-surfacedelta/P07-F04.svg",
        "status": "source-generated conceptual architecture",
        "title": "Matched E1 experiment"
      }
    ],
    "paper": "p07-surfacemind-surfacedelta"
  },
  "evidence_state": "IMPLEMENTED",
  "id": "p07-surfacemind-surfacedelta",
  "institution": "Glyphd Labs",
  "integrity": {
    "claim_ledger_sha256": "0fcd040f25ba3f4d0dcf504d3bb6ac2ee7fc588728153c5c8fb5ccbad49bf3af",
    "paper_mdx_sha256": "138cdb4208cbb01a049a4e39b6ea03428d1753f071ade7e7c640e24f0f54f123"
  },
  "public_disclosure_boundary": "This paper publishes the public-safe graph/delta architecture and the explicit go/no-go experiment. It excludes private source data, unpublished compiler prompts, restricted mobile implementation details, and patent-claim language. SurfaceDelta uplift is not claimed until a versioned matched run is published.",
  "publication_class": "public-safe working paper",
  "publication_url": "https://www.glyphd.com/papers/p07-surfacemind-surfacedelta",
  "revisions": {
    "paper": "p07-surfacemind-surfacedelta",
    "revisions": [
      {
        "date": "2026-07-27",
        "summary": "First synthesized public-safe institute edition compiled from the Glyphd project corpus.",
        "version": "0.1.0"
      }
    ]
  },
  "schema_version": "1.0.0",
  "slug": "p07-surfacemind-surfacedelta",
  "sources": {
    "notes": "Internal source paths are provenance references. Public publication must expose only approved public-safe excerpts or archived public artifacts.",
    "paper": "p07-surfacemind-surfacedelta",
    "sources": [
      {
        "date": "2026-07",
        "id": "P07-S01",
        "path": "bohselecta/surfacemind",
        "title": "SurfaceMind R0 repository and canonical documentation"
      },
      {
        "date": "2026-07-13",
        "id": "P07-S02",
        "path": "Pasted markdown.md",
        "title": "SurfaceMind Pocket / 4B Max-Flex research notes"
      },
      {
        "date": "2026-07-15",
        "id": "P07-S03",
        "path": "ZEKE-4B-MAXFLEX-RUNTIME-MASTER-SPEC-v0.1.md",
        "title": "Zeke 4B Max-Flex Runtime Master Spec"
      },
      {
        "date": "2026-07-24",
        "id": "P07-S04",
        "path": "How_we_can_make_CAS_lightning_fast_for_any_LLM_by_design_or_adaptation.docx",
        "title": "GlyphCAS Design Report"
      },
      {
        "date": "2026-06-18",
        "id": "P07-S05",
        "path": "Workpage_OS_Spec_Pack_v1_1.md",
        "title": "Workpage OS Spec Pack v1.1"
      }
    ],
    "version": "0.1.0"
  },
  "subtitle": "Verified operational surfaces and bounded graph patches for small and large language models",
  "title": "SurfaceMind and SurfaceDelta",
  "version": "0.1.0"
}