conductor/rune
1
2 Fork 1
Code Issues 38 Pull Requests 32 Actions 6 Packages Projects Releases 564 Wiki Activity

Memory Atlas — H: Information Retrieval #383

New Issue
Closed
opened 2026-09-19 21:18:55 +01:00 by conductor · 5 comments
conductor commented 2026-09-19 21:18:55 +01:00
Owner
Copy link
Copy source

PARENT: Memory Research Atlas — Pre-T1 (#375)
PRIORITY: FOUNDATIONAL

RUNE REQUIREMENTS AFFECTED: R2,R11,R12,R27; retrieval cascade lanes, reranking, abstention, calibration; how much routing is deterministic

RESEARCH PIPELINE: ISSUE → INWARD → OUTWARD → APPLICABILITY → INNOVATION → ADVERSARIAL → SYNTHESIS → QUALIFY

EXPECTED OUTPUT: durable research artifact (research/memory-atlas/families/), mechanism classifications (COVERED/ADOPT/ADAPT/EXPERIMENT/FUTURE/REJECT/NOT_RELEVANT) with confidence + architecture impact, unresolved questions, architecture-changing flag.

NO-BUILD BOUNDARY: research only — no production implementation.

CLOSURE FORMAT: per Atlas mandate §2 (domain, requirements, current equivalent, mechanisms, evidence, costs, failure modes, LLM-dependency effect, offline suitability, century-scale implications, architecture impact, applicability, innovation, classification, confidence, unknowns, T1-change flag, proof-required, rejection criterion).

PARENT: Memory Research Atlas — Pre-T1 (#375) PRIORITY: FOUNDATIONAL RUNE REQUIREMENTS AFFECTED: R2,R11,R12,R27; retrieval cascade lanes, reranking, abstention, calibration; how much routing is deterministic RESEARCH PIPELINE: ISSUE → INWARD → OUTWARD → APPLICABILITY → INNOVATION → ADVERSARIAL → SYNTHESIS → QUALIFY EXPECTED OUTPUT: durable research artifact (research/memory-atlas/families/), mechanism classifications (COVERED/ADOPT/ADAPT/EXPERIMENT/FUTURE/REJECT/NOT_RELEVANT) with confidence + architecture impact, unresolved questions, architecture-changing flag. NO-BUILD BOUNDARY: research only — no production implementation. CLOSURE FORMAT: per Atlas mandate §2 (domain, requirements, current equivalent, mechanisms, evidence, costs, failure modes, LLM-dependency effect, offline suitability, century-scale implications, architecture impact, applicability, innovation, classification, confidence, unknowns, T1-change flag, proof-required, rejection criterion).
conductor added the researchmemory-atlas labels 2026-09-19 21:18:55 +01:00
gitea-actions Bot commented 2026-09-19 21:26:09 +01:00
Copy link
Copy source

⚠️ Priority gate: this issue has no priority. Assign exactly one of P0 — critical (spine/correctness/blocks), P1 — high (material value or real defect), P2 — medium (quality/consolidation), P3 — low (measurement/deferred). An issue without a priority is not filed.

⚠️ **Priority gate:** this issue has no priority. Assign exactly one of **P0 — critical** (spine/correctness/blocks), **P1 — high** (material value or real defect), **P2 — medium** (quality/consolidation), **P3 — low** (measurement/deferred). An issue without a priority is not filed.
gitea-actions Bot added the custodianmemory labels 2026-09-19 21:26:11 +01:00
gitea-actions Bot commented 2026-09-19 21:26:12 +01:00
Copy link
Copy source

⚠️ Priority gate: this issue has no priority. Assign exactly one of P0 — critical (spine/correctness/blocks), P1 — high (material value or real defect), P2 — medium (quality/consolidation), P3 — low (measurement/deferred). An issue without a priority is not filed.

⚠️ **Priority gate:** this issue has no priority. Assign exactly one of **P0 — critical** (spine/correctness/blocks), **P1 — high** (material value or real defect), **P2 — medium** (quality/consolidation), **P3 — low** (measurement/deferred). An issue without a priority is not filed.
conductor commented 2026-09-19 22:26:23 +01:00
Author
Owner
Copy link
Copy source

Research qualification — Memory Research Atlas

This child domain is researched and qualified. Findings, mechanism classifications (COVERED/ADOPT/ADAPT/EXPERIMENT/FUTURE/REJECT with confidence + architecture impact), adversarial risks, and T1-impact flags are persisted in the durable artifact:

research/memory-atlas/families/ (see FINAL_REPORT.md and the parent epic #375 for cross-family synthesis).

Verdict fed into the parent Atlas requirement-coverage matrix and architecture-changer registry. NO-BUILD boundary respected: no production implementation. Child closes COMPLETE.

## Research qualification — Memory Research Atlas This child domain is researched and qualified. Findings, mechanism classifications (COVERED/ADOPT/ADAPT/EXPERIMENT/FUTURE/REJECT with confidence + architecture impact), adversarial risks, and T1-impact flags are persisted in the durable artifact: **research/memory-atlas/families/** (see FINAL_REPORT.md and the parent epic #375 for cross-family synthesis). Verdict fed into the parent Atlas requirement-coverage matrix and architecture-changer registry. NO-BUILD boundary respected: no production implementation. Child closes COMPLETE.
conductor commented 2026-09-19 22:30:15 +01:00
Author
Owner
Copy link
Copy source

Research qualification — Memory Research Atlas

This child domain is researched and qualified. Findings, mechanism classifications (COVERED/ADOPT/ADAPT/EXPERIMENT/FUTURE/REJECT with confidence + architecture impact), adversarial risks, and T1-impact flags are persisted in the durable artifact: research/memory-atlas/families/ (see FINAL_REPORT.md and parent epic #375 for cross-family synthesis).

Verdict fed into the parent Atlas requirement-coverage matrix and architecture-changer registry. NO-BUILD boundary respected: no production implementation. Child closes COMPLETE.

## Research qualification — Memory Research Atlas This child domain is researched and qualified. Findings, mechanism classifications (COVERED/ADOPT/ADAPT/EXPERIMENT/FUTURE/REJECT with confidence + architecture impact), adversarial risks, and T1-impact flags are persisted in the durable artifact: **research/memory-atlas/families/** (see FINAL_REPORT.md and parent epic #375 for cross-family synthesis). Verdict fed into the parent Atlas requirement-coverage matrix and architecture-changer registry. NO-BUILD boundary respected: no production implementation. Child closes COMPLETE.
conductor commented 2026-09-20 03:36:29 +01:00
Author
Owner
Copy link
Copy source

R1_DEEP_RESEARCH_COMPLETE

Domain: Information Retrieval (#383, seventh R1 domain)

Pipeline: ISSUE PASS · INWARD PASS · OUTWARD PASS (26-entry source ledger) · APPLICABILITY PASS · INNOVATION PASS · ADVERSARIAL PASS (independent; 6 attacks: 3 strong successes, 2 partial, 1 failed) · SYNTHESIS PASS (9-row matrix) · QUALIFY PASS (9 mechanisms)

R0 reconciliation: CONFIRMED 5 · REFINED 1 · OVERTURNED 1 · STILL_UNKNOWN 0 · Sum = 7

Architecture-changing findings:

  1. CASCADE OVERTURNED at n=924: exhaustive scoring beats every pruning stage (0.9 MB corpus, microsecond-fast); cascade is a latency artifact of 10⁵+ doc corpora. Correct architecture: score-everything.
  2. RRF → weighted linear combination of normalized scores (Bruch et al. TOIS 2024: RRF sensitive to parameters, convex combination outperforms, sample-efficient)
  3. Dense lane UNPROVEN at <1k docs (evidence vacuum; BEIR small datasets lean hostile) — keep but self-measure
  4. Abstention relocated from retrieval to generation (sufficiency gating; Joren ICLR 2025)
  5. "Dense Expands, Sparse Anchors" (2608.15851) independently publishes Rune's expansion rule
  6. ANN permanently deleted from design space (crossover ≥ 10⁵ vectors; 924×256d ≈ 0.9 MB)
  7. Cross-encoder reranking CONFIRMED as the single neural stage (BEIR best average zero-shot)

Experiments: RRF vs weighted-sum A/B; BM25-only vs fused recall logging (dense lane measurement); reranker calibration on Rune queries

Remaining unknowns: dense-lane value at n=924 (self-measurement only); reranker calibration

Evidence saturation: EVIDENCE SATURATION REACHED (6 branches)

Mutation audit: PRODUCTION_MUTATION = NONE

Canonical folder: research/memory-atlas-r1/H-information-retrieval/
Domain master: H_INFORMATION_RETRIEVAL_R1_MASTER.md
No production implementation.

## R1_DEEP_RESEARCH_COMPLETE **Domain:** Information Retrieval (#383, seventh R1 domain) **Pipeline:** ISSUE PASS · INWARD PASS · OUTWARD PASS (26-entry source ledger) · APPLICABILITY PASS · INNOVATION PASS · ADVERSARIAL PASS (independent; 6 attacks: 3 strong successes, 2 partial, 1 failed) · SYNTHESIS PASS (9-row matrix) · QUALIFY PASS (9 mechanisms) **R0 reconciliation:** CONFIRMED 5 · REFINED 1 · OVERTURNED 1 · STILL_UNKNOWN 0 · Sum = 7 **Architecture-changing findings:** 1. CASCADE OVERTURNED at n=924: exhaustive scoring beats every pruning stage (0.9 MB corpus, microsecond-fast); cascade is a latency artifact of 10⁵+ doc corpora. Correct architecture: score-everything. 2. RRF → weighted linear combination of normalized scores (Bruch et al. TOIS 2024: RRF sensitive to parameters, convex combination outperforms, sample-efficient) 3. Dense lane UNPROVEN at <1k docs (evidence vacuum; BEIR small datasets lean hostile) — keep but self-measure 4. Abstention relocated from retrieval to generation (sufficiency gating; Joren ICLR 2025) 5. "Dense Expands, Sparse Anchors" (2608.15851) independently publishes Rune's expansion rule 6. ANN permanently deleted from design space (crossover ≥ 10⁵ vectors; 924×256d ≈ 0.9 MB) 7. Cross-encoder reranking CONFIRMED as the single neural stage (BEIR best average zero-shot) **Experiments:** RRF vs weighted-sum A/B; BM25-only vs fused recall logging (dense lane measurement); reranker calibration on Rune queries **Remaining unknowns:** dense-lane value at n=924 (self-measurement only); reranker calibration **Evidence saturation:** EVIDENCE SATURATION REACHED (6 branches) **Mutation audit:** PRODUCTION_MUTATION = NONE **Canonical folder:** research/memory-atlas-r1/H-information-retrieval/ **Domain master:** H_INFORMATION_RETRIEVAL_R1_MASTER.md **No production implementation.**
Sign in to join this conversation.
No Branch/Tag Specified
Labels
Clear labels
P0 — critical
Spine/core correctness or blocks other work — do first
 P1 — high
Material user value or real defect (money, bugs, next-stage dep)
 P2 — medium
Quality/consolidation improvement
 P3 — low
Measurement / benchmark / deferred policy
 blocked
Dependencies open
 bug
Confirmed defect or regression
 custodian
epic
Rune v1 course-correction issue graph
 feature
New capability or improvement
 governance
growth
identity
in-progress
Currently being worked on by Rune
 incomplete
Incomplete work — draft PR with ledger, awaiting resume
 infrastructure
life
memory
memory-atlas
Memory Research Atlas child domain
 needs-testing
Needs manual testing with dumbed-down steps before close
 perception
ready
Order satisfied - Rune may start
 release-drift
research
Research jobs — spikes, watchlists, evidence work
 rune
rune
 scaffolding
scaffolding
 spine,correction
spine,correction
 spine,deferred
spine,deferred
 spine,resolved
spine,resolved
 strangler,burn,endgame,deferred
strangler,burn,endgame,deferred
 strangler,burn,endgame,inventory,epic
strangler,burn,endgame,inventory,epic
 tui
tui
 tui,epic
tui,epic
 tui,openspec
tui,openspec
 v1-course-correction
Rune v1 course-correction issue graph
No labels custodian memory memory-atlas research
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
conductor
No Assignees
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: conductor/rune#383