WFGY/ProblemMap/Diagnose.md

7.5 KiB
Raw Blame History

🩺 Semantic Failure Diagnostic Sheet

Select the symptom(s) you observe.
Each entry links to the corresponding solution page in the WFGY Problem Map.
🧩 Prefer runnable examples? MVP DemosProblemMap/mvp_demo/README.md


Quick Nav

Problem Map 1.0 · RAG Problem Map 2.0 · Semantic Clinic Index · Retrieval Playbook · Rerankers · Data Contracts · Multilingual Guide · Privacy & Governance


Core 16 failures

# Symptom Problem ID Solution
1 Retriever returns wrong/irrelevant chunks; citations miss expected section #1 Hallucination & Chunk Drift hallucination.md
2 Correct chunks are present, but reasoning is wrong #2 Interpretation Collapse retrieval-collapse.md
3 Multi-step tasks drift off-topic after a few hops #3 Long Reasoning Chains context-drift.md
4 Model answers confidently with made-up facts #4 Bluffing / Overconfidence bluffing.md
5 High cosine similarity but meaning is wrong #5 Semantic ≠ Embedding embedding-vs-semantic.md
6 Logic dead-ends; retries loop or reset nonsense #6 Logic Collapse & Recovery logic-collapse.md
7 Long conversation: model forgets previous context #7 Memory Breaks Across Sessions memory-coherence.md
8 Pipeline is opaque; unable to trace “why this snippet” #8 Debugging is a Black Box retrieval-traceability.md
9 Attention melts; output incoherent or repetitive #9 Entropy Collapse entropy-collapse.md
10 Responses become flat, literal, lose creativity #10 Creative Freeze creative-freeze.md
11 Formal/symbolic prompts break the model #11 Symbolic Collapse symbolic-collapse.md
12 Self-reference / paradox crashes reasoning #12 Philosophical Recursion philosophical-recursion.md
13 Multiple agents overwrite or misalign logic (overview) #13 Multi-Agent Chaos Multi-Agent_Problems.md
14 System runs but outputs nothing; boot order off #14 Bootstrap Ordering bootstrap-ordering.md
15 System never reaches expected state; actions stall #15 Deployment Deadlock deployment-deadlock.md
16 First prod call after deploy crashes / “empty logic” #16 Pre-Deploy Collapse predeploy-collapse.md

Extended patterns (targeted fixes)

Pattern When to use Fix page
Rerankers (ordering control) Recall seems fine but top-k ordering is messy rerankers.md
Retrieval Playbook (end-to-end knobs) You want a guided checklist across retriever params retrieval-playbook.md
Query Parsing Split HyDE/BM25 hybrid worse than single retriever patterns/pattern_query_parsing_split.md
Symbolic Constraint Unlock (SCU) “Who said what” merges across sources; cross-bleed patterns/pattern_symbolic_constraint_unlock.md
Hallucination Re-entry You correct the model, but the wrong claim returns patterns/pattern_hallucination_reentry.md
Vectorstore Fragmentation Some facts cant be retrieved though indexed patterns/pattern_vectorstore_fragmentation.md
Memory Desync Tabs/sessions flip between old/new facts patterns/pattern_memory_desync.md
Bootstrap Deadlock (RAG boot fence) Tools fire before data/index is ready patterns/pattern_bootstrap_deadlock.md
Data Contracts Need a standard schema for snippets/citations data-contracts.md
Multilingual Guide Non-English corpora drift / tokenizer mismatch multilingual-guide.md
Privacy & Governance PII/compliance concerns for traces/logs privacy-and-governance.md

Minimal triage rules

  • Measure first:
    • ΔS(question, retrieved_context) = 1 cosθ
    • High risk if ΔS ≥ 0.60; investigate if 0.400.60 and λ ∈ {←, <>}.
  • Accept when: ΔS ≤ 0.45 · λ stays convergent (→) on ≥3 paraphrases · E_resonance flat.
  • Coverage sanity: retrieved tokens vs target section ≥ 0.70 for direct QA.

👉 Not sure where it fits? Run ΔS / λ_observe first, or use the MVP demos for quick diagnosis: python ProblemMap/mvp_demo/main.py (run from repo root)


🔗 Quick-Start Downloads (60 sec)

Tool Link 3-Step Setup
WFGY 1.0 PDF Engine Paper 1 Download · 2 Upload to your LLM · 3 Ask “Answer using WFGY + <your question>”
TXT OS (plain-text OS) TXTOS.txt 1 Download · 2 Paste into any LLM chat · 3 Type “hello world” — OS boots instantly

Explore More

Layer Page What its for
Proof WFGY Recognition Map External citations, integrations, and ecosystem proof
Engine WFGY 1.0 Original PDF based tension engine
Engine WFGY 2.0 Production tension kernel and math engine for RAG and agents
Engine WFGY 3.0 TXT based Singularity tension engine, 131 S class set
Map Problem Map 1.0 Flagship 16 problem RAG failure checklist and fix map
Map Problem Map 2.0 RAG focused recovery pipeline
Map Problem Map 3.0 Global Debug Card, image as a debug protocol layer
Map Semantic Clinic Symptom to family to exact fix
Map Grandmas Clinic Plain language stories mapped to Problem Map 1.0
Onboarding Starter Village Guided tour for newcomers
App TXT OS TXT semantic OS, fast boot
App Blah Blah Blah Abstract and paradox Q and A built on TXT OS
App Blur Blur Blur Text to image with semantic control
App Blow Blow Blow Reasoning game engine and memory demo

If this repository helped, starring it improves discovery so more builders can find the docs and tools. GitHub Repo stars