Repository navigation
Literature Review Section Crosswalk
Code comments in the main repo cite RESEARCH.md §N. That file is gone: the literature review lives in the Literature Review pages, and this table maps each old section number to its page.
See also the Literature Review overview.
Correction, 2026-08-20: this file previously claimed every RESEARCH.md §N reference in code
comments still points to the same section number on the wiki. That's no longer true, a later wiki
condensation pass replaced every numbered header with a descriptive one, so §14e etc. are not
literal anchors anymore. The content itself wasn't lost, just renamed and shortened. Use this
crosswalk to find where an old section number now lives:
| Old section | Now lives in | Section name |
|---|---|---|
| §1 | Verified Papers | whole page |
| §3 | Leads & Corrections | "Downgraded, do not cite without re verifying" |
| §3b | Leads & Corrections | "Read and rejected" |
| §4 | Leads & Corrections | "Open questions, resolved" |
| §5 | Leads & Corrections | "What's merged into the main repo's docs" |
| §6 | Architecture Synthesis | "A non generative routing layer for delegate_tasks" |
| §7 | Architecture Synthesis | "Comparative survey" |
| §8 | Architecture Synthesis | "RAG reconsideration" |
| §9 / §9a | Architecture Synthesis | "Is DeepDelve's verification architecture novel, or documented prior art?" + "A gap closing follow up" |
| §10 | Hardware & Serving | "Why a structural fix beats a textual warning for recurring citation fabrication" |
| §11 | Hardware & Serving | "Was ROCm the cause of a 9 candidate vLLM disqualification streak?" |
| §12 | Hardware & Serving | "Ollama and llama.cpp tuning for this hardware" |
| §13 / §13a | Hardware & Serving | "Qwen3 think suppression: fixable template gap, or unfixable serving bug?" |
| §14, §14e, §14f | Bake off Findings I | "Ornith-1.0-9B live bake off", plus its "Native backend tool call corruption" and "The clean re test that finally settled it" subsections |
| §15 | Bake off Findings I | "Root cause of a Searcher over fetching for single fact tasks" |
| §16, §17 | Bake off Findings I / II | §16 "Comparative survey extension" (Findings I); §17 "Chasing convergence on the Lisbon/Mexico retest" (Findings II) |
| §18, §18b-f | Bake off Findings II | "The 2026-08-17 seven-fix arc, cross-checked against current literature" |
| §19 | Bake off Findings II | "AgentFloor: a capability-threshold benchmark directly on topic for this project's model search" |
If a specific number/quote/table cited in a code comment needs to be re-verified against the exact
original wording, the pre-condensation version is still in git history: git show 87e3ebb~1:RESEARCH.md.
Design & Plans
History
Model Research
- Model Bake off
- Fine-tuning
- Literature Review
- Small Model Reliability
- Builder Fix Durability
- Harness Design
- Open Items Literature
Reviews & Audits
- Academic Compliance Review
- Comparative Analysis
- Landscape Survey
- Codebase Review
- Report Defect Audit
- Comprehension Diagnostic
- Default Model
- Pipeline Model Comparison
- Writer Redesign Plan
- Writer Role A/B Protocol
- Writer Role A/B Results
Reference