Skip to content

Publish graph/RAG citation navigation promotion report #214

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yvette-carlisle merged 1 commit into from
Jun 19, 2026
Merged

Publish graph/RAG citation navigation promotion report #214

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -267,7 +267,9 @@ provider-backed ELF evidence was required.
comparison blocked; graphify is `wrong_result`; llm-wiki is not_tested; gbrain is
blocked; private and hosted graph/RAG profiles are non_goal. These reports preserve
the smoke and typed non-pass boundaries and do not create an ELF win claim against
graph/RAG strengths.
graph/RAG strengths. Graph/RAG citation/navigation promotion after XY-985 refreshes
this state as 0 pass, 1 wrong_result, 1 incomplete, and 3 blocked, with graphify
evidence-linked output still scoring wrong_result.
- mem0/OpenMemory history follow-up after XY-924 and XY-931: the local OSS mem0
adapter now passes encoded preference correction history, entity-scoped
personalization, local `get_all` export-style readback, and deletion audit history.
Expand Down Expand Up @@ -318,6 +320,7 @@ Detailed evidence and interpretation:
- [Proactive Brief Scoring Report - June 16, 2026](docs/evidence/benchmarking/2026-06-16-proactive-brief-scoring-report.md)
- [Scheduled Memory Task Scoring Report - June 16, 2026](docs/evidence/benchmarking/2026-06-16-scheduled-memory-task-scoring-report.md)
- [Dreaming Competitor-Strength Retest Report - June 17, 2026](docs/evidence/benchmarking/2026-06-17-dreaming-competitor-strength-retest-report.md)
- [Graph/RAG Citation and Navigation Promotion Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-graph-rag-citation-navigation-promotion-report.md)
- [qmd Debug-Ergonomics Dreaming Retest Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-qmd-debug-ergonomics-dreaming-retest-report.md)
- [OpenViking Trajectory Materialization Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-openviking-trajectory-materialization-report.md)
- [Service-Native Dreaming Readback Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-service-native-dreaming-readback-report.md)
Expand Down Expand Up @@ -406,6 +409,7 @@ Detailed comparison, mechanism-level analysis, and source map:
- [Proactive Brief Scoring Report - June 16, 2026](docs/evidence/benchmarking/2026-06-16-proactive-brief-scoring-report.md)
- [Scheduled Memory Task Scoring Report - June 16, 2026](docs/evidence/benchmarking/2026-06-16-scheduled-memory-task-scoring-report.md)
- [Dreaming Competitor-Strength Retest Report - June 17, 2026](docs/evidence/benchmarking/2026-06-17-dreaming-competitor-strength-retest-report.md)
- [Graph/RAG Citation and Navigation Promotion Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-graph-rag-citation-navigation-promotion-report.md)
- [qmd Debug-Ergonomics Dreaming Retest Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-qmd-debug-ergonomics-dreaming-retest-report.md)
- [OpenMemory UI/Export Product Readback Report - June 19, 2026](docs/evidence/benchmarking/2026-06-19-openmemory-ui-export-product-readback-report.md)
- [Live Baseline Benchmark Runbook](docs/runbook/benchmarking/live_baseline_benchmark.md)
Expand Down
168 changes: 168 additions & 0 deletions .../fixtures/report_snapshots/2026-06-19-graph-rag-citation-navigation-promotion-report.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,168 @@
{
"schema": "elf.graph_rag_citation_navigation_promotion_report/v1",
"report_id": "xy-985-graph-rag-citation-navigation-promotion-2026-06-19",
"authority": "XY-985",
"created_at": "2026-06-19T07:17:34Z",
"goal": "Promote graph/RAG citation, navigation, stale-source lint, and knowledge-surface cases only when adapters emit comparable evidence-linked outputs while preserving typed non-pass outcomes.",
"command": {
"command": "cargo make real-world-memory-graph-rag",
"status": "pass",
"report_artifact": "tmp/real-world-memory/graph-rag/report.json",
"markdown_artifact": "tmp/real-world-memory/graph-rag/report.md",
"run_id": "real-world-memory-graph-rag",
"adapter_id": "fixture_graph_rag_external_adapters"
},
"source_baseline": {
"previous_report": "docs/evidence/benchmarking/2026-06-17-dreaming-competitor-strength-retest-report.md",
"previous_graph_rag_report": "docs/evidence/benchmarking/2026-06-11-graph-rag-scored-smoke-adapter-report.md",
"previous_status": "typed_non_pass",
"previous_counts": {
"pass": 0,
"wrong_result": 1,
"incomplete": 1,
"blocked": 3
}
},
"summary": {
"overall_judgment": "unchanged_typed_non_pass",
"broader_graph_rag_parity": "not_proven",
"job_count": 5,
"encoded_suite_count": 3,
"pass": 0,
"wrong_result": 1,
"incomplete": 1,
"blocked": 3,
"not_encoded": 0,
"wrong_result_count": 2,
"regressed_scenario_count": 0,
"evidence_coverage": 0.25,
"source_ref_coverage": 0.25,
"quote_coverage": 0.25,
"knowledge_citation_coverage": 0.667,
"stale_claim_detection": 0.0,
"unsupported_summary_count": 1,
"mean_score": 0.06
},
"scenario_outcomes": [
{
"project": "RAGFlow",
"scenario_id": "reference_chunk_citation_mapping",
"fixture": "apps/elf-eval/fixtures/real_world_external_adapters/graph_rag/ragflow_reference_chunks_blocked.json",
"required_output": "Returned reference chunks must include generated document ids, chunk ids, content, and document metadata mapped to benchmark evidence ids.",
"current_status": "blocked",
"judgment": "unchanged",
"artifact": "tmp/real-world-memory/graph-rag/report.json",
"blocker": "resource_api_setup_and_reference_chunks_missing",
"claim_boundary": "No RAGFlow citation quality or ELF-over-RAGFlow claim is allowed."
},
{
"project": "LightRAG",
"scenario_id": "context_source_reference_mapping",
"fixture": "apps/elf-eval/fixtures/real_world_external_adapters/graph_rag/lightrag_context_sources_incomplete.json",
"required_output": "Context/source export must expose generated file paths, snippets, or references mapped to evidence ids.",
"current_status": "incomplete",
"judgment": "unchanged",
"artifact": "tmp/real-world-memory/graph-rag/report.json",
"blocker": "default_api_export_not_available",
"claim_boundary": "No LightRAG graph/RAG quality claim is allowed until source references map to generated evidence."
},
{
"project": "GraphRAG",
"scenario_id": "output_table_citation_mapping",
"fixture": "apps/elf-eval/fixtures/real_world_external_adapters/graph_rag/graphrag_output_tables_blocked.json",
"required_output": "Output tables must map documents, text units, communities, reports, entities, and relationships to generated evidence ids.",
"current_status": "blocked",
"judgment": "unchanged",
"artifact": "tmp/real-world-memory/graph-rag/report.json",
"blocker": "provider_backed_output_tables_missing",
"claim_boundary": "No GraphRAG citation, synthesis, or navigation claim is allowed without mapped output tables."
},
{
"project": "Graphiti/Zep",
"scenario_id": "temporal_graph_validity_mapping",
"fixture": "apps/elf-eval/fixtures/real_world_external_adapters/graph_rag/graphiti_temporal_validity_blocked.json",
"required_output": "Current and historical graph facts must carry validity windows and evidence ids.",
"current_status": "blocked",
"judgment": "unchanged",
"artifact": "tmp/real-world-memory/graph-rag/report.json",
"blocker": "provider_backed_temporal_graph_output_missing",
"claim_boundary": "Graphiti/Zep remains the temporal-validity reference; no ELF superiority claim is allowed."
},
{
"project": "graphify",
"scenario_id": "graph_report_navigation_lint",
"fixture": "apps/elf-eval/fixtures/real_world_external_adapters/graph_rag/graphify_graph_report_wrong_result.json",
"required_output": "graph.json, source-location sections, unsupported-claim lint, and stale-source lint must all be scored.",
"current_status": "wrong_result",
"judgment": "unchanged",
"artifact": "tmp/real-world-memory/graph-rag/report.json",
"produced_evidence": [
"graphify-derived-report-boundary",
"graphify-graph-summary-output",
"graphify-source-location-output"
],
"blocker": "stale_claim_detection_missing_and_unsupported_summary_present",
"claim_boundary": "graphify has evidence-linked output but remains wrong_result; do not convert this into an ELF win."
},
{
"project": "llm-wiki",
"scenario_id": "wiki_page_citation_lint",
"fixture": null,
"required_output": "Contained page generation must emit cited sections plus stale-source and unsupported-claim lint.",
"current_status": "not_encoded",
"judgment": "unchanged",
"artifact": null,
"blocker": "no_contained_page_materializer",
"claim_boundary": "llm-wiki remains a reference workflow until a contained materializer exists."
},
{
"project": "gbrain",
"scenario_id": "compiled_truth_timeline_export",
"fixture": null,
"required_output": "Docker-local brain repository import must emit compiled-truth or timeline pages with source evidence.",
"current_status": "blocked",
"judgment": "unchanged",
"artifact": null,
"blocker": "docker_local_brain_repo_and_database_setup_missing",
"claim_boundary": "gbrain remains blocked until setup/export readback is proven."
}
],
"improvement_regression_readback": {
"judgment": "unchanged",
"improved": [
"The fresh June 19 report records graphify evidence-linked output and typed blockers in a checked-in XY-985 companion."
],
"unchanged": [
"No graph/RAG scenario moved to pass.",
"RAGFlow, GraphRAG, and Graphiti/Zep remain blocked.",
"LightRAG remains incomplete.",
"graphify remains wrong_result.",
"llm-wiki remains not_encoded and gbrain remains blocked."
],
"regressed": []
},
"claim_boundaries": {
"allowed": [
"The representative graph/RAG command is reproducible and emits typed non-pass outcomes.",
"graphify emits evidence-linked graph/report output but remains wrong_result.",
"The comparison status is unchanged relative to XY-955."
],
"not_allowed": [
"Do not claim graph/RAG parity or broad graph-navigation quality.",
"Do not convert research gates, tiny smokes, blocked setup, incomplete output, or graphify wrong_result into a win.",
"Do not use private providers, hosted services, or unrecorded credentials for this lane."
]
},
"next_optimization_direction": {
"required_fields": [
"ragflow_reference_chunk_ids_and_document_metadata",
"lightrag_context_source_paths_or_snippets",
"graphrag_output_table_rows_with_generated_evidence_ids",
"graphiti_zep_valid_at_invalid_at_evidence_mapping",
"graphify_stale_source_lint_pass",
"llm_wiki_contained_page_materializer",
"gbrain_docker_local_brain_repo_export"
],
"non_goal": "Do not implement broad ELF graph/RAG product features in this benchmark/report lane."
}
}
86 changes: 86 additions & 0 deletions apps/elf-eval/tests/real_world_job_benchmark.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -250,6 +250,10 @@ fn openmemory_ui_export_product_readback_report_json_path() -> Result<PathBuf> {
report_snapshot_path("2026-06-19-openmemory-ui-export-product-readback-report.json")
}

fn graph_rag_citation_navigation_promotion_report_json_path() -> Result<PathBuf> {
report_snapshot_path("2026-06-19-graph-rag-citation-navigation-promotion-report.json")
}

fn openviking_trajectory_materialization_report_markdown_path() -> Result<PathBuf> {
Ok(workspace_root()?
.join("docs")
Expand Down Expand Up @@ -282,6 +286,14 @@ fn openmemory_ui_export_product_readback_report_markdown_path() -> Result<PathBu
.join("2026-06-19-openmemory-ui-export-product-readback-report.md"))
}

fn graph_rag_citation_navigation_promotion_report_markdown_path() -> Result<PathBuf> {
Ok(workspace_root()?
.join("docs")
.join("evidence")
.join("benchmarking")
.join("2026-06-19-graph-rag-citation-navigation-promotion-report.md"))
}

fn live_temporal_reconciliation_report_json_path() -> Result<PathBuf> {
report_snapshot_path("2026-06-16-live-temporal-reconciliation-report.json")
}
Expand Down Expand Up @@ -3505,6 +3517,80 @@ fn openmemory_ui_export_product_recheck_preserves_blocked_boundary() -> Result<(
Ok(())
}

#[test]
fn graph_rag_citation_navigation_promotion_preserves_typed_non_passes() -> Result<()> {
let report = serde_json::from_str::<Value>(&fs::read_to_string(
graph_rag_citation_navigation_promotion_report_json_path()?,
)?)?;
let markdown =
fs::read_to_string(graph_rag_citation_navigation_promotion_report_markdown_path()?)?;
let benchmarking_index = fs::read_to_string(benchmarking_index_path()?)?;
let readme = fs::read_to_string(readme_path()?)?;

assert_eq!(
report.pointer("/schema").and_then(Value::as_str),
Some("elf.graph_rag_citation_navigation_promotion_report/v1")
);
assert_eq!(report.pointer("/authority").and_then(Value::as_str), Some("XY-985"));
assert_eq!(
report.pointer("/command/command").and_then(Value::as_str),
Some("cargo make real-world-memory-graph-rag")
);
assert_eq!(report.pointer("/command/status").and_then(Value::as_str), Some("pass"));
assert_eq!(
report.pointer("/summary/overall_judgment").and_then(Value::as_str),
Some("unchanged_typed_non_pass")
);
assert_eq!(report.pointer("/summary/pass").and_then(Value::as_u64), Some(0));
assert_eq!(report.pointer("/summary/wrong_result").and_then(Value::as_u64), Some(1));
assert_eq!(report.pointer("/summary/incomplete").and_then(Value::as_u64), Some(1));
assert_eq!(report.pointer("/summary/blocked").and_then(Value::as_u64), Some(3));
assert_eq!(report.pointer("/summary/evidence_coverage").and_then(Value::as_f64), Some(0.25));
assert_eq!(
report.pointer("/summary/knowledge_citation_coverage").and_then(Value::as_f64),
Some(0.667)
);

let scenarios = array_at(&report, "/scenario_outcomes")?;
let ragflow = find_by_field(scenarios, "/project", "RAGFlow")?;
let lightrag = find_by_field(scenarios, "/project", "LightRAG")?;
let graphrag = find_by_field(scenarios, "/project", "GraphRAG")?;
let graphiti = find_by_field(scenarios, "/project", "Graphiti/Zep")?;
let graphify = find_by_field(scenarios, "/project", "graphify")?;
let llm_wiki = find_by_field(scenarios, "/project", "llm-wiki")?;
let gbrain = find_by_field(scenarios, "/project", "gbrain")?;

assert_eq!(ragflow.pointer("/current_status").and_then(Value::as_str), Some("blocked"));
assert_eq!(lightrag.pointer("/current_status").and_then(Value::as_str), Some("incomplete"));
assert_eq!(graphrag.pointer("/current_status").and_then(Value::as_str), Some("blocked"));
assert_eq!(graphiti.pointer("/current_status").and_then(Value::as_str), Some("blocked"));
assert_eq!(graphify.pointer("/current_status").and_then(Value::as_str), Some("wrong_result"));
assert_eq!(llm_wiki.pointer("/current_status").and_then(Value::as_str), Some("not_encoded"));
assert_eq!(gbrain.pointer("/current_status").and_then(Value::as_str), Some("blocked"));
assert!(array_contains_str(graphify, "/produced_evidence", "graphify-source-location-output")?);
assert!(array_contains_str(
&report,
"/claim_boundaries/not_allowed",
"Do not claim graph/RAG parity or broad graph-navigation quality."
)?);
assert!(array_contains_str(
&report,
"/next_optimization_direction/required_fields",
"graphrag_output_table_rows_with_generated_evidence_ids"
)?);
assert!(markdown.contains("typed non-pass, no parity claim"));
assert!(
markdown.contains("graphify produces evidence-linked output but still scores wrong_result")
);
assert!(
benchmarking_index.contains("2026-06-19-graph-rag-citation-navigation-promotion-report.md")
);
assert!(readme.contains("Graph/RAG Citation and Navigation Promotion Report - June 19, 2026"));
assert!(readme.contains("Graph/RAG citation/navigation promotion after XY-985"));

Ok(())
}

fn assert_openviking_trajectory_materialization_summary(report: &Value) -> Result<()> {
assert_eq!(
report.pointer("/schema").and_then(Value::as_str),
Expand Down
Loading