ROADMAP.yaml

# ROADMAP.yaml — canonical, generated by `pmat roadmap sync` (MACS-013).
# Do not edit by hand; regenerate. Content hash covers sources, not wall-clock.
generated_at: "2026-07-04T19:38:52.667319257+00:00"  # excluded from content_hash
content_hash: "20add152cfb3dce627c9a82ef3aaad73c5c6c2fd4ba86108d818d83b5c9115b7"
summary:
  blocked: 4
  cancelled: 15
  completed: 275
  planned: 301
items:
  - id: APR-ANTIGRAVITY-INTEGRATION-001
    title: "Track & document Antigravity integration paths (Anthropic-key base URL, Gemini endpoint, community gateways)"
    status: planned
  - id: APR-ANTIGRAVITY-PARITY-001
    title: "Pillar-5 either-harness parity: apr code drivable by Claude Code OR Google Antigravity with prompt parity"
    status: planned
  - id: APR-FORMAT-002
    title: "New task: APR-FORMAT-002"
    status: completed
  - id: APR-GEMINI-PROXY-001
    title: "apr serve gemini — Gemini generateContent provable-contract proxy (the Antigravity wire surface)"
    status: planned
  - id: "APR-PARITY-001: APR/SafeTensors GGUF parity - all QA scripts pass"
    title: "New task: APR-PARITY-001: APR/SafeTensors GGUF parity - all QA scripts pass"
    status: completed
  - id: APR-POKA-001
    title: "New task: APR-POKA-001"
    status: completed
  - id: APR-PUB-001
    title: "New task: APR-PUB-001"
    status: completed
  - id: "APR-TOK-001: Fix tokenizer decoding for coherent text output"
    title: "New task: APR-TOK-001: Fix tokenizer decoding for coherent text output"
    status: completed
  - id: "APR-TRACE-001: Implement tracing features for APR serve path (5 failing tests)"
    title: "New task: APR-TRACE-001: Implement tracing features for APR serve path (5 failing tests)"
    status: completed
  - id: APR-VERIFY-001
    title: "New task: APR-VERIFY-001"
    status: completed
  - id: APRENDER-METAL-001
    title: "Metal Backend Integration: AMD GPU Tensor Ops via Intel Mac"
    status: cancelled
  - id: "Add APR-SPEC to book and verify examples"
    title: "New task: Add APR-SPEC to book and verify examples"
    status: completed
  - id: "Add contract-driven property tests from provable-contracts"
    title: "New task: Add contract-driven property tests from provable-contracts"
    status: completed
  - id: "Apply code review safety fixes: Add AprenderError enum and bounds checking to SIMD intrinsics"
    title: "New task: Apply code review safety fixes: Add AprenderError enum and bounds checking to SIMD intrinsics"
    status: completed
  - id: BEAT-OLLAMA-DECODE-CI-001
    title: "Wire beat_ollama_decode_throughput_speed into a GPU-nightly (developer-verified -> CI-enforced)"
    status: planned
  - id: C-04-hardcoded-eos-chat
    title: "New task: C-04-hardcoded-eos-chat"
    status: completed
  - id: CUDA-CI-NIGHTLY-001
    title: "CUDA nightly CI lane — enforce the GPU QLoRA training-correctness falsifiers"
    status: completed
  - id: "Commit integration specification and proptest configuration"
    title: "New task: Commit integration specification and proptest configuration"
    status: completed
  - id: "Fix 4 failing Bayesian doctests"
    title: "New task: Fix 4 failing Bayesian doctests"
    status: completed
  - id: "Fix mdbook test failures - Phase 10 (remaining example chapters)"
    title: "New task: Fix mdbook test failures - Phase 10 (remaining example chapters)"
    status: completed
  - id: "Fix mdbook test failures - Phase 11 (pca-iris.md remaining failures)"
    title: "New task: Fix mdbook test failures - Phase 11 (pca-iris.md remaining failures)"
    status: completed
  - id: "Fix mdbook test failures - Phase 4 (20 chapters remaining)"
    title: "New task: Fix mdbook test failures - Phase 4 (20 chapters remaining)"
    status: completed
  - id: "Fix mdbook test failures - Phase 5 (~17 chapters remaining)"
    title: "New task: Fix mdbook test failures - Phase 5 (~17 chapters remaining)"
    status: completed
  - id: "Fix mdbook test failures - Phase 6 (~15 chapters remaining)"
    title: "New task: Fix mdbook test failures - Phase 6 (~15 chapters remaining)"
    status: completed
  - id: "Fix mdbook test failures - Phase 7 (pca.md + classification-metrics.md)"
    title: "New task: Fix mdbook test failures - Phase 7 (pca.md + classification-metrics.md)"
    status: completed
  - id: "Fix mdbook test failures - Phase 8 (cross-validation.md + graph-algorithms.md)"
    title: "New task: Fix mdbook test failures - Phase 8 (cross-validation.md + graph-algorithms.md)"
    status: completed
  - id: "Fix mdbook test failures - Phase 9 (descriptive-statistics.md + gradient-descent.md)"
    title: "New task: Fix mdbook test failures - Phase 9 (descriptive-statistics.md + gradient-descent.md)"
    status: completed
  - id: "Fix mdbook test failures in Bayesian inference chapters"
    title: "New task: Fix mdbook test failures in Bayesian inference chapters"
    status: completed
  - id: "Fix remaining mdbook test failures - Phase 3 (21 chapters remaining)"
    title: "New task: Fix remaining mdbook test failures - Phase 3 (21 chapters remaining)"
    status: completed
  - id: "GGUF-SERVE-001: Validate apr serve works for all GGUF model sizes (0.5B, 1B, 1.5B, 7B, 32B)"
    title: "GGUF-SERVE-001: Validate apr serve works for all GGUF model sizes"
    status: completed
  - id: GH-10
    title: "Add Graph and Statistics Examples (cargo examples + book chapters)"
    status: completed
  - id: GH-11
    title: "Add Graph and Statistics Book Chapters"
    status: completed
  - id: GH-13
    title: "Implement PCA (Principal Component Analysis) for Dimensionality Reduction"
    status: completed
  - id: GH-15
    title: "New task: GH-15"
    status: completed
  - id: GH-16
    title: "New task: GH-16"
    status: completed
  - id: GH-17
    title: "New task: GH-17"
    status: completed
  - id: GH-18
    title: "New task: GH-18"
    status: completed
  - id: GH-189
    title: "APR chat special tokens fix (BpeTokenizer)"
    status: completed
  - id: GH-19
    title: "New task: GH-19"
    status: completed
  - id: GH-20
    title: "New task: GH-20"
    status: completed
  - id: GH-202
    title: "New task: GH-202"
    status: completed
  - id: GH-205
    title: "New task: GH-205"
    status: completed
  - id: GH-207
    title: "New task: GH-207"
    status: completed
  - id: GH-208
    title: "New task: GH-208"
    status: completed
  - id: GH-21
    title: "New task: GH-21"
    status: completed
  - id: GH-219
    title: "New task: GH-219"
    status: completed
  - id: GH-22
    title: "New task: GH-22"
    status: completed
  - id: GH-23
    title: "[TOP 10] Implement K-Nearest Neighbors (kNN) Classifier"
    status: completed
  - id: "GH-234/235/236: lm_head quantization skip, GPT-2 hidden_dim metadata, GGUF export architecture"
    title: "New task: GH-234/235/236: lm_head quantization skip, GPT-2 hidden_dim metadata, GGUF export architecture"
    status: completed
  - id: GH-237
    title: "New task: GH-237"
    status: completed
  - id: GH-242
    title: "apr eval rejects APR format + GGUF fused_matmul type error"
    status: completed
  - id: GH-243
    title: "feat: apr quantize — unified quantization pipeline with GGUF/AWQ/GPTQ output"
    status: completed
  - id: GH-244
    title: "feat: apr finetune — LoRA/QLoRA fine-tuning with adapter upload to Hub"
    status: completed
  - id: GH-246
    title: "feat: apr export — MLX, CoreML, and OpenVINO format support"
    status: completed
  - id: GH-247
    title: "feat: apr prune + apr distill — structured pruning and knowledge distillation pipeline"
    status: completed
  - id: GH-248
    title: "bug: --json flag ignored on list, eval, tree, debug subcommands"
    status: completed
  - id: GH-249
    title: "bug: apr inspect --json returns empty architecture field"
    status: completed
  - id: GH-250
    title: "bug: apr run --json produces completely wrong tokens (32/32 mismatch)"
    status: completed
  - id: GH-251
    title: "bug: apr validate output format changed — dict with top-level keys instead of checks list"
    status: completed
  - id: GH-252
    title: "bug: apr lint exits 5 on warnings-only (0 errors)"
    status: completed
  - id: GH-253
    title: "bug: apr check (selftest) fails with GH-187 embedding warning"
    status: completed
  - id: GH-254
    title: "bug: apr bench reports 0.0 tok/s throughput on all models"
    status: completed
  - id: GH-256
    title: "bug: apr diff reports structural diffs between int4 and int8 variants of same model"
    status: completed
  - id: GH-26
    title: "[TOP 10] Implement Gradient Boosting Machine (GBM)"
    status: completed
  - id: GH-279-4
    title: "GH-279: Wire QkNorm into GPU forward path + fix format_parity sharded SafeTensors"
    status: completed
  - id: GH-28
    title: "New task: GH-28"
    status: completed
  - id: GH-280
    title: "GH-280: Kernel capability gate — contract-driven GPU admission control"
    status: completed
  - id: GH-282
    title: "fix: apr serve GPU failures — missing make_current() in APR CUDA path"
    status: completed
  - id: GH-29
    title: "New task: GH-29"
    status: completed
  - id: GH-30
    title: "New task: GH-30"
    status: completed
  - id: GH-306
    title: "New task: GH-306"
    status: completed
  - id: GH-31
    title: "New task: GH-31"
    status: completed
  - id: GH-32
    title: "New task: GH-32"
    status: completed
  - id: GH-324
    title: "New task: GH-324"
    status: completed
  - id: GH-33
    title: "New task: GH-33"
    status: completed
  - id: GH-336
    title: "fix(P2): benchmark silently swallows generate() errors, reports 0 tok/s"
    status: completed
  - id: GH-337
    title: "fix(P2): chat server silently degrades to byte-level decode on tokenizer error"
    status: completed
  - id: GH-338
    title: "fix(P2): probar silently ignores corrupted msgpack metadata"
    status: completed
  - id: GH-339
    title: "fix(P2): chat template failure silently falls back to raw prompt"
    status: completed
  - id: GH-34
    title: "New task: GH-34"
    status: completed
  - id: GH-340
    title: "fix(P3): safetensors.rs placeholder generation produces fake tokens"
    status: completed
  - id: GH-341
    title: "chore: publish safety gate + package bloat cleanup (28MB savings)"
    status: completed
  - id: GH-35
    title: "New task: GH-35"
    status: completed
  - id: GH-36
    title: "New task: GH-36"
    status: completed
  - id: GH-366
    title: "New task: GH-366"
    status: completed
  - id: GH-37
    title: "New task: GH-37"
    status: completed
  - id: GH-38
    title: "New task: GH-38"
    status: completed
  - id: GH-39
    title: "New task: GH-39"
    status: completed
  - id: GH-40
    title: "New task: GH-40"
    status: completed
  - id: GH-41
    title: "New task: GH-41"
    status: completed
  - id: GH-42
    title: "New task: GH-42"
    status: completed
  - id: GH-43
    title: "New task: GH-43"
    status: completed
  - id: GH-44
    title: "New task: GH-44"
    status: completed
  - id: GH-45
    title: "New task: GH-45"
    status: completed
  - id: GH-479
    title: "New task: GH-479"
    status: completed
  - id: GH-55
    title: "New task: GH-55"
    status: completed
  - id: GH-71
    title: "Implement Content-Based Recommender with HNSW (Phase 1)"
    status: completed
  - id: GH-73
    title: "feat: Model evaluation framework for .apr models with retraining support"
    status: completed
  - id: GH-80
    title: "feat(metaheuristics): Implement derivative-free global optimization for aprender-contrib"
    status: completed
  - id: GH-9
    title: "Implement graph algorithms and descriptive statistics"
    status: completed
  - id: "Implement ARIMA (Auto-Regressive Integrated Moving Average)"
    title: "New task: Implement ARIMA (Auto-Regressive Integrated Moving Average)"
    status: completed
  - id: "Implement Bayesian Linear Regression (analytical posterior)"
    title: "New task: Implement Bayesian Linear Regression (analytical posterior)"
    status: completed
  - id: "Implement Bayesian Logistic Regression (Laplace approximation)"
    title: "New task: Implement Bayesian Logistic Regression (Laplace approximation)"
    status: completed
  - id: "Implement Bayesian Model Selection (Bayes Factors, DIC, WAIC)"
    title: "New task: Implement Bayesian Model Selection (Bayes Factors, DIC, WAIC)"
    status: completed
  - id: "Implement Covariance and Correlation Matrices"
    title: "New task: Implement Covariance and Correlation Matrices"
    status: completed
  - id: "Implement Generalized Linear Models (GLM) - Poisson, Gamma, Binomial families"
    title: "New task: Implement Generalized Linear Models (GLM) - Poisson, Gamma, Binomial families"
    status: completed
  - id: "Implement Independent Component Analysis (ICA)"
    title: "New task: Implement Independent Component Analysis (ICA)"
    status: completed
  - id: "Implement Statistical Hypothesis Tests (t-test, chi-square, ANOVA)"
    title: "New task: Implement Statistical Hypothesis Tests (t-test, chi-square, ANOVA)"
    status: completed
  - id: "Implement apr-cli TUI command"
    title: "New task: Implement apr-cli TUI command"
    status: completed
  - id: "Implement document similarity (cosine, Jaccard, edit distance)"
    title: "New task: Implement document similarity (cosine, Jaccard, edit distance)"
    status: completed
  - id: "Implement extractive summarization (TextRank, TF-IDF)"
    title: "New task: Implement extractive summarization (TextRank, TF-IDF)"
    status: completed
  - id: "Implement pattern-based entity extraction (emails, URLs, keywords)"
    title: "New task: Implement pattern-based entity extraction (emails, URLs, keywords)"
    status: completed
  - id: "LAYOUT-001: Fix Q4K kernel layout mismatch - use row-major kernel for APR/GGUF"
    title: "New task: LAYOUT-001: Fix Q4K kernel layout mismatch - use row-major kernel for APR/GGUF"
    status: completed
  - id: PILLAR1-001
    title: "cross_val_score (model_selection)"
    status: planned
  - id: PILLAR1-002
    title: "KNeighborsRegressor (neighbors)"
    status: planned
  - id: PILLAR1-003
    title: "GaussianNB Estimator trait impl (naive_bayes)"
    status: planned
  - id: PILLAR1-004
    title: "KNearestNeighbors Estimator trait impl (neighbors)"
    status: planned
  - id: PILLAR1-005
    title: "LogisticRegression Estimator trait impl (linear_model)"
    status: completed
  - id: PILLAR1-006
    title: "DecisionTreeClassifier Estimator trait impl (tree)"
    status: completed
  - id: PILLAR1-007
    title: "DecisionTreeRegressor Estimator trait impl (tree)"
    status: planned
  - id: PILLAR1-008
    title: "RandomForestRegressor Estimator trait impl (ensemble)"
    status: planned
  - id: PILLAR1-009
    title: "GradientBoostingClassifier Estimator trait impl (ensemble)"
    status: planned
  - id: PILLAR1-010
    title: "LinearSVM Estimator trait impl (classification.svm)"
    status: planned
  - id: PILLAR1-011
    title: "score() method on LinearSVM (classification.svm)"
    status: planned
  - id: PILLAR1-012
    title: "predict_proba for DecisionTreeClassifier (tree)"
    status: planned
  - id: PILLAR1-013
    title: "min_samples_split for DecisionTreeClassifier (tree)"
    status: planned
  - id: PILLAR1-014
    title: "min_samples_leaf for DecisionTreeClassifier (tree)"
    status: planned
  - id: PILLAR1-015
    title: "PCA explained_variance attributes (decomposition)"
    status: planned
  - id: PILLAR1-016
    title: "ICA Transformer trait impl (decomposition)"
    status: planned
  - id: PILLAR1-017
    title: "LocalOutlierFactor UnsupervisedEstimator trait (cluster)"
    status: planned
  - id: PILLAR1-018
    title: "IsolationForest UnsupervisedEstimator trait (cluster)"
    status: planned
  - id: PILLAR1-019
    title: "MaxAbsScaler (preprocessing)"
    status: planned
  - id: PILLAR1-020
    title: "Normalizer (preprocessing)"
    status: planned
  - id: PILLAR1-021
    title: "LabelEncoder (preprocessing)"
    status: planned
  - id: PILLAR1-022
    title: "cohen_kappa_score (metrics)"
    status: planned
  - id: PILLAR1-023
    title: "hamming_loss (metrics)"
    status: planned
  - id: PILLAR1-024
    title: "jaccard_score (metrics)"
    status: planned
  - id: PILLAR1-025
    title: "fbeta_score (metrics)"
    status: planned
  - id: PILLAR1-026
    title: "explained_variance_score (metrics)"
    status: planned
  - id: PILLAR1-027
    title: "max_error (metrics)"
    status: planned
  - id: PILLAR1-028
    title: "median_absolute_error (metrics)"
    status: planned
  - id: PILLAR1-029
    title: "mean_squared_log_error (metrics)"
    status: planned
  - id: PILLAR1-030
    title: "mean_absolute_percentage_error (metrics)"
    status: planned
  - id: PILLAR1-031
    title: "davies_bouldin_score (metrics)"
    status: planned
  - id: PILLAR1-032
    title: "calinski_harabasz_score (cluster)"
    status: planned
  - id: PILLAR1-033
    title: "adjusted_rand_score (metrics)"
    status: planned
  - id: PILLAR1-034
    title: "silhouette_samples (cluster)"
    status: planned
  - id: PILLAR1-035
    title: "min_impurity_decrease parameter (tree)"
    status: planned
  - id: PILLAR1-036
    title: "KNN weights=distance parameter (neighbors)"
    status: planned
  - id: PILLAR1-037
    title: "make_pipeline (pipeline_compose)"
    status: planned
  - id: PILLAR1-038
    title: "ShuffleSplit (model_selection)"
    status: planned
  - id: PILLAR1-039
    title: "LeaveOneOut (model_selection)"
    status: planned
  - id: PILLAR1-040
    title: "Naive Bayes alpha (Laplace smoothing) parameter (naive_bayes)"
    status: planned
  - id: PILLAR1-041
    title: "GridSearchCV (model_selection)"
    status: planned
  - id: PILLAR1-042
    title: "GradientBoostingRegressor (ensemble)"
    status: planned
  - id: PILLAR1-043
    title: "OneHotEncoder (preprocessing)"
    status: planned
  - id: PILLAR1-044
    title: "TruncatedSVD (decomposition)"
    status: planned
  - id: PILLAR1-045
    title: "NMF (decomposition)"
    status: planned
  - id: PILLAR1-046
    title: "MultinomialNB (naive_bayes)"
    status: planned
  - id: PILLAR1-047
    title: "Pipeline (ML composition) (pipeline_compose)"
    status: planned
  - id: PILLAR1-048
    title: "learning_curve (model_selection)"
    status: planned
  - id: PILLAR1-049
    title: "validation_curve (model_selection)"
    status: planned
  - id: PILLAR1-050
    title: "RandomizedSearchCV (model_selection)"
    status: planned
  - id: PILLAR1-051
    title: "SGDClassifier (linear_model)"
    status: planned
  - id: PILLAR1-052
    title: "SGDRegressor (linear_model)"
    status: planned
  - id: PILLAR1-053
    title: "Kernel classes (RBF, Poly, Sigmoid, Linear) (classification.svm)"
    status: planned
  - id: PILLAR1-054
    title: "Multi-class SVM (OneVsRest) (classification.svm)"
    status: planned
  - id: PILLAR1-055
    title: "MiniBatchKMeans (cluster)"
    status: planned
  - id: PILLAR1-056
    title: "max_features parameter for RandomForest (tree)"
    status: planned
  - id: PILLAR1-057
    title: "class_weight=balanced parameter (tree)"
    status: planned
  - id: PILLAR1-058
    title: "AdaBoostClassifier (ensemble)"
    status: planned
  - id: PILLAR1-059
    title: "VotingClassifier (ensemble)"
    status: planned
  - id: PILLAR1-060
    title: "OrdinalEncoder (preprocessing)"
    status: planned
  - id: PILLAR1-061
    title: "PolynomialFeatures (preprocessing)"
    status: planned
  - id: PILLAR1-062
    title: "FactorAnalysis (decomposition)"
    status: planned
  - id: PILLAR1-063
    title: "BernoulliNB (naive_bayes)"
    status: planned
  - id: PILLAR1-064
    title: "ComplementNB (naive_bayes)"
    status: planned
  - id: PILLAR1-065
    title: "roc_curve (metrics)"
    status: planned
  - id: PILLAR1-066
    title: "precision_recall_curve (metrics)"
    status: planned
  - id: PILLAR1-067
    title: "cross_validate_unsupervised (cluster)"
    status: planned
  - id: PILLAR1-068
    title: "NearestNeighbors (generic query interface) (neighbors)"
    status: planned
  - id: PILLAR1-069
    title: "RadiusNeighborsClassifier (neighbors)"
    status: planned
  - id: PILLAR1-070
    title: "ColumnTransformer (pipeline_compose)"
    status: planned
  - id: PMAT-081
    title: "FORMAT-PARITY-001: Implement SafeTensors and APR inference"
    status: completed
  - id: PMAT-082
    title: "APR-CLI-RELEASE-029: Release apr-cli v0.2.10 to crates.io"
    status: completed
  - id: PMAT-083
    title: "PMAT-COMPLY-001"
    status: completed
  - id: PMAT-084
    title: "FILE-HEALTH-001: Refactor large files to improve code health"
    status: completed
  - id: PMAT-085
    title: "FILE-HEALTH-002: Split optim/mod.rs into submodules"
    status: completed
  - id: PMAT-086
    title: "SERVE-TRACE-001: Implement X-Trace-Level header in apr serve"
    status: completed
  - id: PMAT-087
    title: "SERVE-STREAM-001: Fix SSE streaming not visible to OpenAI SDK clients"
    status: completed
  - id: PMAT-088
    title: "SERVE-CHAT-001: Fix prompt injection in chat completions (extra Human/Assistant turns)"
    status: completed
  - id: PMAT-089
    title: "F-CLI-006-REGRESSION: apr check subcommand missing from binary"
    status: completed
  - id: PMAT-090
    title: "PAR-303-REGRESSION: 0.5B model produces incoherent gibberish output"
    status: completed
  - id: PMAT-091
    title: "REALIZE-TRACE-001: X-Trace-Level header not working with apr serve delegation"
    status: completed
  - id: PMAT-092
    title: "PMAT-092: Fix empty stop_tokens in GPU chat completions path causing multi-turn loop"
    status: completed
  - id: PMAT-093
    title: "PAR-301: Expand apr serve to support APR and SafeTensors formats"
    status: completed
  - id: PMAT-094
    title: "FIVE-WHYS-ST-001: SafeTensors inference produces garbage output"
    status: completed
  - id: PMAT-095
    title: "PERF-ST-001: SafeTensors performance parity with GGUF"
    status: completed
  - id: PMAT-096
    title: "PMAT-098: APR format performance parity with GGUF and SafeTensors"
    status: completed
  - id: PMAT-097
    title: "REALIZAR-SEND-001: GpuModel Send/Sync bounds fix for apr-cli cuda feature"
    status: completed
  - id: PMAT-098
    title: "PMAT-QA-PROTOCOL-001: Implement comprehensive QA testing protocol"
    status: completed
  - id: PMAT-099
    title: "PMAT-110: APR CUDA KV Cache - Implement proper KV cache for AprV2ModelCuda generation"
    status: completed
  - id: "PMAT-099: APR format parity - fix AprTransformer config reading"
    title: "New task: PMAT-099: APR format parity - fix AprTransformer config reading"
    status: completed
  - id: "PMAT-103-SIMD: AVX2 SIMD optimization for Q4K matmul to achieve 5+ tok/s"
    title: "New task: PMAT-103-SIMD: AVX2 SIMD optimization for Q4K matmul to achieve 5+ tok/s"
    status: completed
  - id: PMAT-110
    title: "New task: PMAT-110"
    status: cancelled
  - id: PMAT-111
    title: "PMAT-114: SafeTensors→APR inference fix"
    status: completed
  - id: PMAT-112
    title: "FALSIFY-100: Run 100-point Popperian falsification checklist"
    status: completed
  - id: PMAT-113
    title: "PMAT-115: Fix SafeTensors GPU silent fallback (Jidoka violation)"
    status: completed
  - id: PMAT-114
    title: "PMAT-116: Implement SafeTensors GPU inference"
    status: completed
  - id: PMAT-115
    title: "PMAT-117: Implement SafeTensors GPU inference (from spec)"
    status: completed
  - id: PMAT-116
    title: "New task: PMAT-116"
    status: completed
  - id: PMAT-117
    title: "Fix SATD violations in apr-cli (Toyota Way compliance)"
    status: completed
  - id: PMAT-118
    title: "F-PERF-042: GPU > 2x CPU throughput verification"
    status: completed
  - id: PMAT-119
    title: "F-QUAL-032: Argmax parity GGUF vs SafeTensors QA gate"
    status: completed
  - id: PMAT-120
    title: "DEFECT: SafeTensors GPU chat produces garbage (Andon)"
    status: completed
  - id: PMAT-121
    title: "FALSIFY-SPEC-001: Systematic falsification of qwen2.5-coder-showcase-demo.md claims"
    status: completed
  - id: PMAT-122
    title: "PMAT-123: Continue systematic falsification testing"
    status: completed
  - id: PMAT-124
    title: "New task: PMAT-124"
    status: completed
  - id: "PMAT-127: Fix F-APR-GGUF - APR from GGUF produces garbage"
    title: "FIXED: PMAT-127 APR from GGUF produces garbage (dtype mismatch)"
    status: completed
  - id: "PMAT-129: Fix F-SAFETENSORS-GPU - apr run should support SafeTensors GPU"
    title: "New task: PMAT-129: Fix F-SAFETENSORS-GPU - apr run should support SafeTensors GPU"
    status: completed
  - id: "PMAT-130: Fix F-Q4_0 - GGUF Q4_0 dequantization produces garbage"
    title: "New task: PMAT-130: Fix F-Q4_0 - GGUF Q4_0 dequantization produces garbage"
    status: completed
  - id: "PMAT-131: Fix F-UX-033 - SafeTensors CPU should display backend type"
    title: "New task: PMAT-131: Fix F-UX-033 - SafeTensors CPU should display backend type"
    status: completed
  - id: PMAT-190
    title: "GH-201: Eliminate hot-path unwrap() calls"
    status: completed
  - id: PMAT-191
    title: "GH-202: Remove Unknown Error from explain.rs"
    status: completed
  - id: PMAT-192
    title: "GH-203: Cross-architecture determinism validation"
    status: completed
  - id: PMAT-193
    title: "GH-204: Prompt injection sanitization"
    status: completed
  - id: PMAT-194
    title: "GH-205: Load testing infrastructure"
    status: completed
  - id: PMAT-195
    title: "GH-198: apr pull SafeTensors missing companion files (tokenizer.json, config.json)"
    status: completed
  - id: PMAT-196
    title: "GGUF-EXPORT-001"
    status: completed
  - id: PMAT-197
    title: "PMAT-210: Fix BUG-1 GGUF Export from SafeTensors (CRASH)"
    status: completed
  - id: PMAT-198
    title: "PMAT-201: Complete Per-Tensor Statistical Fingerprints"
    status: completed
  - id: PMAT-199
    title: "PMAT-202: Tensor Statistics Validation"
    status: completed
  - id: PMAT-200
    title: "PMAT-203: Golden Output Embedding"
    status: completed
  - id: PMAT-201
    title: "PMAT-204: Tensor Distribution Tags"
    status: completed
  - id: PMAT-202
    title: "PMAT-205: Sharding-Aware Placement"
    status: cancelled
  - id: PMAT-203
    title: "BUG-RUN-001: Fix benchmark token count mismatch in apr run"
    status: completed
  - id: PMAT-204
    title: "STACK-DRIFT-001: Update downstream crates for aprender 0.25.1"
    status: completed
  - id: PMAT-205
    title: "GH-202-FIX-002: Debug realizar APR loader tensor shapes"
    status: completed
  - id: PMAT-206
    title: "GH-202-FIX-003: E2E tensor comparison test for GGUF vs APR"
    status: completed
  - id: PMAT-207
    title: "GH-202-FIX-004: realizar E2E test comparing GGUF vs APR inference"
    status: completed
  - id: PMAT-208
    title: "GH-202: LAYOUT-002 conversion fidelity fix"
    status: completed
  - id: PMAT-209
    title: "GH-200: Qwen2.5-Coder dev-tier QA qualification"
    status: completed
  - id: PMAT-210
    title: "GH-199: APR 1.5B dequantize and GPU issues"
    status: completed
  - id: PMAT-211
    title: "GH-192: APR inference performance (0.5 vs 270 tok/s)"
    status: completed
  - id: PMAT-212
    title: "LAYOUT-CONTRACT-001: Implement tensor layout contract validation tests"
    status: completed
  - id: PMAT-213
    title: "LAYOUT-CONTRACT-002: Finalize spec documentation and playbook integration"
    status: completed
  - id: PMAT-214
    title: "SPEC-FINAL-001: Complete tensor layout contract specification"
    status: completed
  - id: PMAT-215
    title: "GH-208: APR Inference - Layer Weight Investigation"
    status: cancelled
  - id: PMAT-216
    title: "GH-208-GPU: Fix GPU inference garbage output (Five Whys root cause analysis)"
    status: cancelled
  - id: PMAT-217
    title: "GH-205: F16 SafeTensors Passthrough Fix"
    status: completed
  - id: PMAT-218
    title: "PMAT-240: Create model family YAMLs (qwen2, llama, whisper, bert)"
    status: completed
  - id: PMAT-219
    title: "PMAT-241: ModelFamily trait + ModelFamilyConfig"
    status: completed
  - id: PMAT-220
    title: "PMAT-242: YAML contract loader"
    status: completed
  - id: PMAT-221
    title: "PMAT-243: Clippy column-major bans"
    status: completed
  - id: PMAT-222
    title: "PMAT-244: apr oracle local file mode"
    status: completed
  - id: PMAT-223
    title: "PMAT-245: apr oracle HuggingFace API mode"
    status: completed
  - id: PMAT-224
    title: "PMAT-246: apr oracle --family contract mode"
    status: completed
  - id: PMAT-225
    title: "PMAT-247: Playbook certification cross-reference"
    status: completed
  - id: PMAT-226
    title: "PMAT-248: PhantomData<Layout> on Validated* types"
    status: completed
  - id: PMAT-227
    title: "PMAT-249: AprTransformer migration to Validated* fields"
    status: completed
  - id: PMAT-228
    title: "PMAT-250: build.rs YAML-to-Rust code generation"
    status: completed
  - id: PMAT-229
    title: "PMAT-251: Generic AprTransformer<F: ModelFamily>"
    status: cancelled
  - id: PMAT-230
    title: "PMAT-253: Additional model family YAMLs"
    status: completed
  - id: PMAT-231
    title: "GH-254: Verify F2-VALIDATION BOS fix and update showcase spec FALSIFIED gates"
    status: completed
  - id: PMAT-232
    title: "GH-255: 7B GGUF GPU produces garbage output (CPU correct)"
    status: completed
  - id: PMAT-233
    title: "PMAT-260: Investigate 23.7 GB RSS for 7B Q4K (4 GB on disk) — 6x model size"
    status: completed
  - id: PMAT-234
    title: "PMAT-234: F-GT-001 enforce-provenance flag + F-GT-002 mixed quant warning"
    status: completed
  - id: PMAT-235
    title: "PMAT-PROFILE-A: Formal verification + code quality for A score"
    status: completed
  - id: PMAT-243
    title: "Fix 16 showcase spec falsification bugs (Round 51)"
    status: completed
  - id: PMAT-244
    title: "Split 3 files >2000 lines for File Health compliance"
    status: completed
  - id: PMAT-245
    title: "Fix CB-500 string byte indexing warnings"
    status: completed
  - id: PMAT-246
    title: "Fix CB-900 broken markdown links in book"
    status: completed
  - id: PMAT-257
    title: "GH-259: apr flow — add GGUF/SafeTensors support via Rosetta dispatch"
    status: completed
  - id: PMAT-258
    title: "GH-260: apr diff — fix cross-format APR parse failure"
    status: completed
  - id: PMAT-259
    title: "apr hex --stats silently drops stats for BF16 tensors"
    status: completed
  - id: PMAT-260
    title: "GH-279/280/281: Entrenar delegation API gaps (Loss Send+Sync, cross_entropy_loss, per-class metrics)"
    status: completed
  - id: PMAT-261
    title: "Qwen3-8B: Performance Parity Pipeline"
    status: completed
  - id: PMAT-262
    title: "GH-277/278: Fix GGUF export compatibility with llama.cpp"
    status: completed
  - id: PMAT-283
    title: "GH-279-SEAL-001: Seal OwnedQuantizedModel fields pub(crate)"
    status: completed
  - id: PMAT-284
    title: "GH-279-SEAL-002: Validated input wrapper for GpuModel::from_apr_weights"
    status: completed
  - id: PMAT-285
    title: "GH-279-SEAL-003: Consolidate transpose to trueno-quant single source"
    status: completed
  - id: PMAT-286
    title: "UCBD-001: Create provable-contracts-macros proc macro crate"
    status: completed
  - id: PMAT-287
    title: "UCBD-002: Add build.rs binding verification to realizar"
    status: completed
  - id: PMAT-288
    title: "UCBD-003: Delete aprender GGUF reader/dequant dead code"
    status: completed
  - id: PMAT-289
    title: "UCBD-004: Delete aprender SafeTensors loader dead code"
    status: cancelled
  - id: PMAT-290
    title: "UCBD-005: Delete realizar duplicate helpers"
    status: cancelled
  - id: PMAT-291
    title: "UCBD-006: Wire entrenar to use realizar for model loading"
    status: cancelled
  - id: PMAT-292
    title: "UCBD-007: Implement all 49 not_implemented bindings"
    status: cancelled
  - id: PMAT-293
    title: "New task: PMAT-293"
    status: completed
  - id: PMAT-295
    title: "New task: PMAT-295"
    status: completed
  - id: PMAT-296
    title: "GH-279 GAP-1: SafeTensors import completeness gate is conditional"
    status: completed
  - id: PMAT-297
    title: "GH-279 GAP-2: Export path has zero architecture completeness checking"
    status: completed
  - id: PMAT-298
    title: "GH-279 GAP-3: validate_model_load_basic skips role checking on CPU paths"
    status: completed
  - id: PMAT-299
    title: "GH-279 GAP-4: CPU AprTransformer has no architecture completeness gate"
    status: completed
  - id: PMAT-325
    title: "FALSIFY-E6: realizar GPU/GGUF embedding path skips ValidatedEmbedding data quality gates"
    status: blocked
  - id: PMAT-326
    title: "FALSIFY-E7: entrenar saves/loads embeddings without any data quality validation"
    status: blocked
  - id: PMAT-327
    title: "FALSIFY-E8: realizar OOB token_id bounds check is insufficient in GPU forward paths"
    status: blocked
  - id: PMAT-328
    title: "FALSIFY-L1: lm_head matmul dimension validation at call sites (§2.1.2)"
    status: completed
  - id: PMAT-329
    title: "FALSIFY-L2: entrenar lm_head save/load has no shape or quality validation (§2.1.2)"
    status: blocked
  - id: PMAT-330
    title: "PMAT-332: ValidatedVector accepts zero-length norm vectors — no minimum length gate. Zero-length norm weight produces NaN in LayerNorm division. Gate 1 only checks data.len()==expected_len, needs data.len()>0 guard."
    status: completed
  - id: PMAT-331
    title: "PMAT-333: FFN gate_proj/up_proj shape symmetry not enforced — ValidatedWeight accepts any [out,in] without verifying gate==up shape or down==[hidden,intermediate] reversal. SwiGLU requires gate_proj.shape==up_proj.shape."
    status: completed
  - id: PMAT-332
    title: "PMAT-334: FALSIFY-007 dispatch exhaustiveness test only scans 2 of 6 dispatch sites listed in tensor-layout-v1.yaml. Missing: brick/dispatch.rs, quantize/dispatch.rs, layers/attention.rs, gpu/scheduler.rs. A catch-all in any unscanned file would go undetected."
    status: completed
  - id: PMAT-333
    title: "PMAT-335: Q4_1 and Q5_0 formats in GPU dispatch (gemv_dispatch.rs) but missing from ALL_FORMAT_IDS registry and FALSIFY-QDOT contract tests. Orphaned formats could silently dispatch to wrong kernel without contract coverage."
    status: completed
  - id: PMAT-334
    title: "PMAT-336: FALSIFY-QDOT-006 missing — gap in numbering between QDOT-005 and QDOT-007. No test verifies row-major-only constraint from quantized-dot-product-v1.yaml enforcement.row_major_only."
    status: completed
  - id: PMAT-335
    title: "PMAT-337: Model-family YAML contract falsification tests (FALSIFY-MF-001..013)"
    status: completed
  - id: PMAT-336
    title: "PMAT-338: Create special-tokens-registry-v1.yaml + FALSIFY-ST-001..006"
    status: completed
  - id: PMAT-337
    title: "FT-DBC-001: Fine-tuning provable Design by Contract parity"
    status: completed
  - id: PMAT-342
    title: "APR checkpoint spec v1.2.0 — full command matrix and falsification"
    status: completed
  - id: PMAT-367
    title: "fix: TensorDType::from_u8 missing GGML Q4_0/Q5_0/Q5_1 dtype mappings (GH-375)"
    status: completed
  - id: PMAT-368
    title: "fix: apr-qa dim-smoke G0 config.json resolution from HF cache (GH-360)"
    status: completed
  - id: PMAT-369
    title: "fix: sharded SafeTensors symlink resolution (GH-363)"
    status: completed
  - id: PMAT-370
    title: "fix: BPE tokenizer O(n²) greedy rescan → priority queue (GH-378)"
    status: completed
  - id: PMAT-371
    title: "Fix nightly CI: patch alimentar path deps for standalone build"
    status: completed
  - id: PMAT-480
    title: "SPEC-CLI: Implement --trace and --profile global flags on apr run/chat/serve"
    status: completed
  - id: PMAT-481
    title: "SPEC-PROBAR: Implement apr probar golden snapshot command"
    status: completed
  - id: PMAT-482
    title: "SPEC-BRICK: Implement BrickTracer auto-escalation in apr cbtop"
    status: completed
  - id: PMAT-483
    title: "SPEC-CONTRACT: Add provable-contracts equivalence obligations for all 5 compute backends"
    status: completed
  - id: PMAT-484
    title: "SPEC-PROBAR-CI: Wire apr probar into tier2/tier3 quality gates"
    status: completed
  - id: PMAT-485
    title: "SPEC-TRACE-OTLP: Add OTLP export to apr serve for distributed tracing"
    status: completed
  - id: PMAT-486
    title: "SPEC-TRAIN-PROFILE: Add --profile and --trace flags to apr train/finetune"
    status: completed
  - id: PMAT-487
    title: "SPEC-DIRTY: Commit pre-existing dirty files with fixes"
    status: completed
  - id: PMAT-488
    title: "SPEC-LAYER2-DISPATCH: Add apr backend select/list commands for Layer 2 GPU dispatch"
    status: completed
  - id: PMAT-489
    title: "BOOK: Add 7 missing chapters to SUMMARY.md"
    status: completed
  - id: PMAT-490
    title: "TEST: Un-ignore 5 time-based tests with fake clock"
    status: completed
  - id: PMAT-491
    title: "TEST: Fix ignored ADMM consensus test"
    status: completed
  - id: PMAT-492
    title: "FIX: cargo install apr-cli — entrenar xa_t compile error"
    status: completed
  - id: PMAT-493
    title: "CONTRACT-GAP: Create 5 highest-priority missing YAML contracts for orphan annotations"
    status: completed
  - id: PMAT-494
    title: "CONTRACT-GAP: Implement FALSIFY-BE-001..003 for compute-backend-equivalence"
    status: completed
  - id: PMAT-495
    title: "CONTRACT-GAP: Implement FALSIFY-NORM-001..003 for normalization-kernel"
    status: completed
  - id: PMAT-496
    title: "SF-002: Thread 7 sampling params through apr run dispatch chain"
    status: completed
  - id: PMAT-497
    title: "APR-BOOK-001: Scaffold 20 chapter examples"
    status: planned
  - id: PMAT-498
    title: "APR-BOOK-002: Create 20 chapter contract YAMLs"
    status: planned
  - id: PMAT-499
    title: "APR-BOOK-003: Create book_contracts integration test"
    status: planned
  - id: PMAT-500
    title: "APR-BOOK-004: Falsify all chapters end-to-end"
    status: planned
  - id: PMAT-501
    title: "APR-BOOK-005: Archive 6 SHOULD-MERGE repos with redirects"
    status: planned
  - id: PMAT-502
    title: "APR-BOOK-006: Archive 24 STALE repos"
    status: planned
  - id: PMAT-503
    title: "APR-BOOK-007: Tag all 205 repos with category topics"
    status: planned
  - id: PMAT-504
    title: "APR-BOOK-008: Archive 18 LEGACY-LIBRARY repos"
    status: planned
  - id: PMAT-505
    title: "APR-BOOK-009: Provable contracts for 21 ACTIVE-TOOL repos"
    status: planned
  - id: PMAT-506
    title: "APR-BOOK-010: Provable contracts for 13 GROUND-TRUTH corpora"
    status: planned
  - id: PMAT-507
    title: "APR-BOOK-011: Convert 7 hollow examples to real API exercises"
    status: planned
  - id: PMAT-508
    title: "APR-BOOK-012: Write Part I chapters (1-3) as markdown prose"
    status: planned
  - id: PMAT-509
    title: "APR-BOOK-013: Provable-contract-only book schema — eliminate all muda"
    status: planned
  - id: PMAT-510
    title: "APR-BOOK-014: PCU contracts for 229 remaining pages — iterative falsification"
    status: planned
  - id: PMAT-511
    title: "APR-BOOK-015: POC benchmark chapters — contract-first vs candle/llama.cpp/ollama/vllm"
    status: planned
  - id: PMAT-512
    title: "APR-BOOK-016: Switch-from chapters — PyTorch/Ollama/unsloth/ndarray migration guides"
    status: planned
  - id: PMAT-513
    title: "APR-BOOK-017: Close proof chain leaks L2-L7, add CI enforcement"
    status: planned
  - id: PMAT-514
    title: "APR-MCP-KAIZEN: Continuous drift sweep on apr-mcp-server-spec.md"
    status: planned
  - id: PMAT-515
    title: "CRUX gap: CRUX-A-08 — Custom mirror / HF_ENDPOINT"
    status: planned
  - id: PMAT-516
    title: "CRUX gap: CRUX-A-10 — VRAM-aware quant auto-select"
    status: planned
  - id: PMAT-517
    title: "CRUX gap: CRUX-A-11 — apr cp copy model with new tag"
    status: planned
  - id: PMAT-518
    title: "CRUX gap: CRUX-A-12 — apr ps list running models"
    status: planned
  - id: PMAT-519
    title: "CRUX gap: CRUX-A-13 — apr stop unload model from VRAM"
    status: planned
  - id: PMAT-520
    title: "CRUX gap: CRUX-A-14 — Pull from S3/GCS/Azure URL"
    status: planned
  - id: PMAT-521
    title: "CRUX gap: CRUX-A-16 — Modelfile recipe custom system prompt"
    status: planned
  - id: PMAT-522
    title: "CRUX gap: CRUX-A-17 — Manifest signing cosign/sigstore"
    status: planned
  - id: PMAT-523
    title: "CRUX gap: CRUX-A-21 — Shared model cache across users"
    status: planned
  - id: PMAT-524
    title: "CRUX gap: CRUX-A-22 — Disk-quota enforcement on registry"
    status: planned
  - id: PMAT-525
    title: "CRUX gap: CRUX-A-23 — Model search Hub + local cache"
    status: planned
  - id: PMAT-526
    title: "CRUX gap: CRUX-A-25 — Garbage-collect unused models"
    status: planned
  - id: PMAT-527
    title: "CRUX gap: CRUX-B-02 — GGUF → Safetensors for PEFT"
    status: planned
  - id: PMAT-528
    title: "CRUX gap: CRUX-B-07 — imatrix calibration"
    status: planned
  - id: PMAT-529
    title: "CRUX gap: CRUX-B-08 — AWQ quantization"
    status: planned
  - id: PMAT-530
    title: "CRUX gap: CRUX-B-09 — GPTQ quantization"
    status: planned
  - id: PMAT-531
    title: "CRUX gap: CRUX-B-10 — BitsAndBytes NF4 4-bit"
    status: planned
  - id: PMAT-532
    title: "CRUX gap: CRUX-B-11 — FP8 quantization H100+"
    status: planned
  - id: PMAT-533
    title: "CRUX gap: CRUX-B-15 — IQ-quants IQ3_XXS / IQ2_S"
    status: planned
  - id: PMAT-534
    title: "CRUX gap: CRUX-B-16 — Per-tensor layer-quant override"
    status: planned
  - id: PMAT-535
    title: "CRUX gap: CRUX-B-17 — Quantization-aware training (QAT)"
    status: planned
  - id: PMAT-536
    title: "CRUX gap: CRUX-B-18 — Calibration dataset loader"
    status: planned
  - id: PMAT-537
    title: "CRUX gap: CRUX-C-04 — Ollama /api/chat"
    status: planned
  - id: PMAT-538
    title: "CRUX gap: CRUX-C-09 — Speculative decoding draft model"
    status: planned
  - id: PMAT-539
    title: "CRUX gap: CRUX-C-11 — OpenAI tool-use function calling"
    status: planned
  - id: PMAT-540
    title: "CRUX gap: CRUX-C-12 — Multi-modal vision input LLaVA"
    status: planned
  - id: PMAT-541
    title: "CRUX gap: CRUX-C-13 — /v1/embeddings endpoint"
    status: planned
  - id: PMAT-542
    title: "CRUX gap: CRUX-C-15 — Tensor + pipeline parallelism"
    status: planned
  - id: PMAT-543
    title: "CRUX gap: CRUX-C-16 — LoRA hotswap at runtime"
    status: planned
  - id: PMAT-544
    title: "CRUX gap: CRUX-C-17 — Multi-LoRA serving N adapters"
    status: planned
  - id: PMAT-545
    title: "CRUX gap: CRUX-C-21 — Mirostat sampling"
    status: planned
  - id: PMAT-546
    title: "CRUX gap: CRUX-C-22 — Typical sampling"
    status: planned
  - id: PMAT-547
    title: "CRUX gap: CRUX-C-23 — DRY sampling"
    status: planned
  - id: PMAT-548
    title: "CRUX gap: CRUX-C-24 — Beam search decoding"
    status: planned
  - id: PMAT-549
    title: "CRUX gap: CRUX-C-29 — NUMA-aware memory placement"
    status: planned
  - id: PMAT-550
    title: "CRUX gap: CRUX-C-30 — KV cache quantization Q8_0/Q4_0"
    status: planned
  - id: PMAT-551
    title: "CRUX gap: CRUX-D-03 — QLoRA 4-bit + LoRA"
    status: planned
  - id: PMAT-552
    title: "CRUX gap: CRUX-D-04 — DPO preference training"
    status: planned
  - id: PMAT-553
    title: "CRUX gap: CRUX-D-11 — DDP multi-GPU single node"
    status: planned
  - id: PMAT-554
    title: "CRUX gap: CRUX-D-12 — FSDP / ZeRO sharding"
    status: planned
  - id: PMAT-555
    title: "CRUX gap: CRUX-D-17 — ORPO / KTO / IPO alignment"
    status: planned
  - id: PMAT-556
    title: "CRUX gap: CRUX-D-18 — Reward-model training"
    status: planned
  - id: PMAT-557
    title: "CRUX gap: CRUX-D-19 — PPO reinforcement learning"
    status: planned
  - id: PMAT-558
    title: "CRUX gap: CRUX-D-20 — RoPE frequency scaling during SFT"
    status: planned
  - id: PMAT-559
    title: "CRUX gap: CRUX-D-24 — Activation checkpointing tags"
    status: planned
  - id: PMAT-560
    title: "CRUX gap: CRUX-D-26 — LR finder range test"
    status: planned
  - id: PMAT-561
    title: "CRUX gap: CRUX-D-27 — Hyperparameter sweep ray/optuna"
    status: planned
  - id: PMAT-562
    title: "CRUX gap: CRUX-D-28 — Adapter fusion combine N LoRAs"
    status: planned
  - id: PMAT-563
    title: "CRUX gap: CRUX-D-29 — Pre-compiled dataset cache"
    status: planned
  - id: PMAT-564
    title: "CRUX gap: CRUX-D-30 — Multi-task training loop"
    status: planned
  - id: PMAT-565
    title: "CRUX gap: CRUX-D-31 — TensorBoard / wandb integration"
    status: planned
  - id: PMAT-566
    title: "CRUX gap: CRUX-D-33 — Distributed optimizer ZeRO-1"
    status: planned
  - id: PMAT-567
    title: "CRUX gap: CRUX-D-34 — Deepspeed config loader"
    status: planned
  - id: PMAT-568
    title: "CRUX gap: CRUX-D-35 — accelerate launch wrapper"
    status: planned
  - id: PMAT-569
    title: "CRUX gap: CRUX-E-02 — Perplexity on held-out corpus"
    status: planned
  - id: PMAT-570
    title: "CRUX gap: CRUX-E-03 — lm-eval-harness tasks"
    status: planned
  - id: PMAT-571
    title: "CRUX gap: CRUX-E-07 — Latency P50/P95/P99 under load"
    status: planned
  - id: PMAT-572
    title: "CRUX gap: CRUX-E-10 — Hallucination / drift detector"
    status: planned
  - id: PMAT-573
    title: "CRUX gap: CRUX-E-11 — MT-Bench / arena judge eval"
    status: planned
  - id: PMAT-574
    title: "CRUX gap: CRUX-E-12 — BBH/MMLU/HellaSwag per-task"
    status: planned
  - id: PMAT-575
    title: "CRUX gap: CRUX-E-13 — RULER long-context eval"
    status: planned
  - id: PMAT-576
    title: "CRUX gap: CRUX-E-14 — Needle-in-haystack recall"
    status: planned
  - id: PMAT-577
    title: "CRUX gap: CRUX-E-19 — Perplexity per quant bit-budget"
    status: planned
  - id: PMAT-578
    title: "CRUX gap: CRUX-E-20 — KL divergence vs FP16 baseline"
    status: planned
  - id: PMAT-579
    title: "CRUX gap: CRUX-E-21 — Bias/toxicity eval harness"
    status: planned
  - id: PMAT-580
    title: "CRUX gap: CRUX-E-22 — Code-eval sandbox Docker runner"
    status: planned
  - id: PMAT-581
    title: "CRUX gap: CRUX-E-23 — Tool-use benchmark BFCL"
    status: planned
  - id: PMAT-582
    title: "CRUX gap: CRUX-E-24 — RAG eval RAGAS/TruLens"
    status: planned
  - id: PMAT-583
    title: "CRUX gap: CRUX-E-25 — Vision-language benchmark harness"
    status: planned
  - id: PMAT-584
    title: "CRUX gap: CRUX-F-06 — KV-cache utilization timeline"
    status: planned
  - id: PMAT-585
    title: "CRUX gap: CRUX-F-07 — GPU memory timeline Chrome trace"
    status: planned
  - id: PMAT-586
    title: "CRUX gap: CRUX-F-09 — Gradient-norm telemetry per step"
    status: planned
  - id: PMAT-587
    title: "CRUX gap: CRUX-F-11 — NaN/Inf detector in activations"
    status: planned
  - id: PMAT-588
    title: "CRUX gap: CRUX-F-13 — CUDA OOM postmortem report"
    status: planned
  - id: PMAT-589
    title: "CRUX gap: CRUX-F-14 — Deadlock/hang detector stack dump"
    status: planned
  - id: PMAT-590
    title: "CRUX gap: CRUX-F-15 — NCCL failure diagnostics"
    status: planned
  - id: PMAT-591
    title: "CRUX gap: CRUX-F-17 — Attention pattern visualization"
    status: planned
  - id: PMAT-592
    title: "CRUX gap: CRUX-F-18 — Token embedding UMAP 2D"
    status: planned
  - id: PMAT-593
    title: "CRUX gap: CRUX-F-19 — apr explain token selection"
    status: planned
  - id: PMAT-594
    title: "CRUX gap: CRUX-G-04 — Auto-generate README with examples"
    status: planned
  - id: PMAT-595
    title: "CRUX gap: CRUX-G-07 — SemVer tag on apr publish"
    status: planned
  - id: PMAT-596
    title: "CRUX gap: CRUX-G-11 — Ollama push to remote registry"
    status: planned
  - id: PMAT-597
    title: "CRUX gap: CRUX-G-14 — License metadata validator"
    status: planned
  - id: PMAT-598
    title: "CRUX gap: CRUX-G-15 — Duplicate-artifact detector pre-push"
    status: planned
  - id: PMAT-599
    title: "CRUX gap: CRUX-H-03 — Packing for efficient SFT"
    status: planned
  - id: PMAT-600
    title: "CRUX gap: CRUX-H-06 — Streaming datasets no RAM mat"
    status: planned
  - id: PMAT-601
    title: "CRUX gap: CRUX-H-12 — ImageFolder loader"
    status: planned
  - id: PMAT-602
    title: "CRUX gap: CRUX-H-13 — Audio dataset loader WAV/FLAC"
    status: planned
  - id: PMAT-603
    title: "CRUX gap: CRUX-H-14 — Dataset mixing per-source weights"
    status: planned
  - id: PMAT-604
    title: "CRUX gap: CRUX-H-16 — Tokenizer-aware length bucketing"
    status: planned
  - id: PMAT-605
    title: "CRUX gap: CRUX-H-17 — Contrastive pair sampler for CLIP"
    status: planned
  - id: PMAT-606
    title: "CRUX gap: CRUX-H-18 — Negative sampling for ranking"
    status: planned
  - id: PMAT-607
    title: "CRUX gap: CRUX-H-19 — Synthetic data generation via LLM"
    status: planned
  - id: PMAT-608
    title: "CRUX gap: CRUX-H-20 — PII redaction in datasets"
    status: planned
  - id: PMAT-609
    title: "CRUX gap: CRUX-H-21 — Row-level dedup MinHash"
    status: planned
  - id: PMAT-610
    title: "CRUX gap: CRUX-I-04 — Ollama function calling"
    status: planned
  - id: PMAT-611
    title: "CRUX gap: CRUX-I-06 — ReAct agent loop + stop conditions"
    status: planned
  - id: PMAT-612
    title: "CRUX gap: CRUX-I-07 — Claude Agent SDK compatibility"
    status: planned
  - id: PMAT-613
    title: "CRUX gap: CRUX-I-09 — Parallel tool-calls one turn"
    status: planned
  - id: PMAT-614
    title: "CRUX gap: CRUX-I-12 — GBNF grammar from JSON-schema"
    status: planned
  - id: PMAT-615
    title: "CRUX gap: CRUX-I-13 — MCP resource provider"
    status: planned
  - id: PMAT-616
    title: "CRUX gap: CRUX-I-14 — MCP prompt provider"
    status: planned
  - id: PMAT-617
    title: "CRUX gap: CRUX-I-15 — Agent memory plugin interface"
    status: planned
  - id: PMAT-618
    title: "CRUX gap: CRUX-I-16 — Guardrails output filter pipeline"
    status: planned
  - id: PMAT-619
    title: "CRUX gap: CRUX-J-01 — CLIP zero-shot classification"
    status: planned
  - id: PMAT-620
    title: "CRUX gap: CRUX-J-02 — Joint text+image embedding"
    status: planned
  - id: PMAT-621
    title: "CRUX gap: CRUX-J-03 — CLIP retrieval cosine search"
    status: planned
  - id: PMAT-622
    title: "CRUX gap: CRUX-J-04 — Visual feature extraction"
    status: planned
  - id: PMAT-623
    title: "CRUX gap: CRUX-J-05 — LAION WebDataset tar loading"
    status: planned
  - id: PMAT-624
    title: "CRUX gap: CRUX-J-06 — SigLIP model support"
    status: planned
  - id: PMAT-625
    title: "CRUX gap: CRUX-J-07 — EVA-CLIP model support"
    status: planned
  - id: PMAT-626
    title: "CRUX gap: CRUX-J-08 — CLIP contrastive training loop"
    status: planned
  - id: PMAT-627
    title: "CRUX gap: CRUX-J-09 — Image preprocessing resize/normalize"
    status: planned
  - id: PMAT-628
    title: "CRUX gap: CRUX-J-10 — Text tokenizer for CLIP BPE"
    status: planned
  - id: PMAT-629
    title: "CRUX gap: CRUX-J-11 — OpenCLIP checkpoint loader"
    status: planned
  - id: PMAT-630
    title: "CRUX gap: CRUX-J-12 — Image-text pair dataloader"
    status: planned
  - id: PMAT-631
    title: "CRUX gap: CRUX-J-13 — Video-CLIP temporal pooling"
    status: planned
  - id: PMAT-632
    title: "CRUX gap: CRUX-J-14 — Audio-CLIP encoder"
    status: planned
  - id: PMAT-633
    title: "CRUX gap: CRUX-J-15 — CLIP fine-tune on new concept"
    status: planned
  - id: PMAT-634
    title: "CRUX gap: CRUX-J-16 — Distill large-CLIP → small-CLIP"
    status: planned
  - id: PMAT-635
    title: "CRUX gap: CRUX-J-17 — Zero-shot detection OWL-ViT"
    status: planned
  - id: PMAT-636
    title: "CRUX gap: CRUX-J-18 — Multilingual CLIP"
    status: planned
  - id: PMAT-637
    title: "CRUX gap: CRUX-J-19 — Export CLIP → ONNX / GGUF"
    status: planned
  - id: PMAT-638
    title: "CRUX gap: CRUX-J-20 — CLIP eval ImageNet zero-shot"
    status: planned
  - id: PMAT-639
    title: "CRUX gap: CRUX-K-02 — Ollama Python SDK test suite"
    status: planned
  - id: PMAT-640
    title: "CRUX gap: CRUX-K-05 — apr ui Gradio web UI"
    status: planned
  - id: PMAT-641
    title: "CRUX gap: CRUX-K-07 — Prometheus /metrics endpoint"
    status: planned
  - id: PMAT-642
    title: "CRUX gap: CRUX-K-08 — OpenTelemetry traces"
    status: planned
  - id: PMAT-643
    title: "CRUX gap: CRUX-K-11 — Modelfile DSL parser (Ollama)"
    status: planned
  - id: PMAT-644
    title: "CRUX gap: CRUX-K-12 — VSCode extension for apr"
    status: planned
  - id: PMAT-645
    title: "CRUX gap: CRUX-K-13 — Docker image apr-serve:latest"
    status: planned
  - id: PMAT-646
    title: "CRUX gap: CRUX-K-14 — systemd unit for apr serve"
    status: planned
  - id: PMAT-647
    title: "CRUX gap: CRUX-K-15 — Kubernetes Helm chart"
    status: planned
  - id: PMAT-648
    title: "CRUX gap: CRUX-K-16 — Triton Inference Server backend"
    status: planned
  - id: PMAT-649
    title: "CRUX gap: CRUX-K-17 — NVIDIA Dynamo integration"
    status: planned
  - id: PMAT-650
    title: "CRUX gap: CRUX-K-18 — ONNX Runtime backend"
    status: planned
  - id: PMAT-651
    title: "CRUX gap: CRUX-K-19 — CoreML export (Apple Silicon)"
    status: planned
  - id: PMAT-652
    title: "CRUX gap: CRUX-K-20 — TensorRT-LLM export"
    status: planned
  - id: PMAT-653
    title: "CRUX gap: CRUX-K-21 — MLX backend (Apple Silicon native)"
    status: planned
  - id: PMAT-654
    title: "CRUX gap: CRUX-A-04 — Include/exclude file globs"
    status: planned
  - id: PMAT-655
    title: "CRUX gap: CRUX-L-01 — Load pre-built HF kernel by hf://kernels-community/<name>"
    status: planned
  - id: PMAT-656
    title: "CRUX gap: CRUX-L-02 — Drop-in flash-attn2 HF kernel behind aprender FA trait"
    status: planned
  - id: PMAT-657
    title: "CRUX gap: CRUX-L-03 — Drop-in flash-attn3 HF kernel (sm_90a / sm_100)"
    status: planned
  - id: PMAT-658
    title: "CRUX gap: CRUX-L-04 — rmsnorm fused kernel behind LayerNorm trait"
    status: planned
  - id: PMAT-659
    title: "CRUX gap: CRUX-L-05 — rotary fused RoPE-apply kernel"
    status: planned
  - id: PMAT-660
    title: "CRUX gap: CRUX-L-06 — paged-attention KV-cache kernel (vLLM-compat)"
    status: planned
  - id: PMAT-661
    title: "CRUX gap: CRUX-L-07 — fp8-fbgemm matmul path for H100+"
    status: planned
  - id: PMAT-662
    title: "CRUX gap: CRUX-L-08 — quantization-bitsandbytes NF4 via HF kernel"
    status: planned
  - id: PMAT-663
    title: "CRUX gap: CRUX-L-09 — quantization-gptq fused dequant+matmul"
    status: planned
  - id: PMAT-664
    title: "CRUX gap: CRUX-L-10 — liger-kernels fused CE-loss for training"
    status: planned
  - id: PMAT-665
    title: "CRUX gap: CRUX-L-11 — megablocks MoE routing kernel"
    status: planned
  - id: PMAT-666
    title: "CRUX gap: CRUX-L-12 — punica-sgmv multi-LoRA fused SGMV"
    status: planned
  - id: PMAT-667
    title: "CRUX gap: CRUX-L-13 — activation fused SwiGLU/GeGLU kernel"
    status: planned
  - id: PMAT-668
    title: "CRUX gap: CRUX-L-14 — mamba-ssm state-space kernel"
    status: planned
  - id: PMAT-669
    title: "CRUX gap: CRUX-L-15 — Kernel version-pinning + SHA-verified load"
    status: planned
  - id: PMAT-670
    title: "CRUX gap: CRUX-M-01 — Gate-1 byte-identical vs safetensors ground truth"
    status: planned
  - id: PMAT-671
    title: "CRUX gap: CRUX-M-02 — Gate-2 tensor-stats parity (min/max/mean/std, per-tensor)"
    status: planned
  - id: PMAT-672
    title: "CRUX gap: CRUX-M-04 — Gate-4 cross-format parity APR ↔ GGUF ↔ safetensors"
    status: planned
  - id: PMAT-673
    title: "CRUX gap: CRUX-M-05 — Gate-5 tokenizer round-trip with NFC normalization"
    status: planned
  - id: PMAT-674
    title: "CRUX gap: CRUX-M-06 — Gate-6 chat-template rendering equivalence"
    status: planned
  - id: PMAT-675
    title: "CRUX gap: CRUX-M-07 — Gate-7 pass@1 on canary suite with confidence interval"
    status: planned
  - id: PMAT-676
    title: "CRUX gap: CRUX-M-08 — Gate-8 5-Whys root-cause generator on failure"
    status: planned
  - id: PMAT-677
    title: "CRUX gap: CRUX-M-09 — Property-based falsifier ≥ 1000 fuzz cases per gate"
    status: planned
  - id: PMAT-678
    title: "CRUX gap: CRUX-M-10 — Upstream-fix enforcement (reject workarounds)"
    status: planned
  - id: PMAT-679
    title: "P0-I: Verify P0-G+P0-H end-to-end on fresh checkpoint"
    status: planned
  - id: PMAT-680
    title: "P0-J: Chinchilla gate — warning → hard blocker"
    status: planned
  - id: PMAT-681
    title: "P2-C: Widen corpus to >2B tokens (highest EV)"
    status: completed
  - id: PMAT-682
    title: "P2-A2: Same-corpus longer run (DOWNGRADED — fallback only)"
    status: planned
  - id: PMAT-683
    title: "P2-D: True distillation from MODEL-1 (apr distill)"
    status: planned
  - id: PMAT-684
    title: "P1-B: HumanEval pass@1 on best checkpoint (DEFERRED)"
    status: planned
  - id: PMAT-685
    title: "P1-C: Python validity 100 prompts (DEFERRED)"
    status: planned
  - id: PMAT-686
    title: "P3-A: apr inspect --quality ≥ 90 on best checkpoint"
    status: planned
  - id: PMAT-687
    title: "P3-B: apr lint zero High severity on best checkpoint"
    status: planned
  - id: PMAT-688
    title: "P3-C: Publish to HuggingFace as paiml/albor-370m-v1"
    status: planned
  - id: PMAT-689
    title: "P3-D: Post-publish QA + /dogfood verdict"
    status: planned
  - id: PMAT-690
    title: "P0-K NEW: apr convert must stamp hf_architecture + tokenizer + merges into imported APRs"
    status: planned
  - id: PMAT-691
    title: "Distillation Phase 2 — KD-wired student forward+backward"
    status: planned
  - id: PMAT-692
    title: "Per-position (full-sequence) knowledge distillation for aprender-train-distill"
    status: completed
  - id: PMAT-693
    title: "Sharded GGUF pull + auto-merge — run multi-part GGUFs end-to-end (closes #1893)"
    status: completed
  - id: PMAT-694
    title: "BF16 (ggml type 30) GGUF loader support"
    status: completed
  - id: PMAT-695
    title: "APR->GGUF export rejects AprQ8 instead of corrupt-relabeling as Q8_0 (import/export symmetry)"
    status: completed
  - id: PMAT-696
    title: "Fix corrupt Q2_K dequantization (185/256 elements wrong) — format + inference paths"
    status: completed
  - id: PMAT-697
    title: "Correct ggml dtype names 26-30 (BF16 was reported as IQ1_M in apr tensors/inspect)"
    status: completed
  - id: PMAT-698
    title: "Hard-refuse unsupported architectures instead of silent llama fallback (replace FALSIFY-TNAME-005)"
    status: planned
  - id: PMAT-710
    title: "Unsloth parity: compete with Unsloth on small-model LoRA/QLoRA fine-tune + distillation (pure Rust)"
    status: planned
  - id: PMAT-711
    title: "QLoRA 4-bit + LoRA fine-tune (apr finetune --qlora --bits 4) — CRUX-D QLoRA (currently MISSING)"
    status: planned
  - id: PMAT-712
    title: "Complete LoRA-merge + export-to-GGUF (Unsloth save_pretrained_merged equivalent) — CRUX-D partial->supported"
    status: planned
  - id: PMAT-713
    title: "End-to-end apr finetune Unsloth-parity single-command UX (load -> QLoRA -> train -> merge -> export -> publish)"
    status: planned
  - id: PMAT-714
    title: "Gradient checkpointing for memory (CRUX-D partial) — Unsloth low-VRAM wedge"
    status: planned
  - id: PMAT-715
    title: "Adapter fusion (combine N LoRAs) + promote Unsloth to tracked training-throughput competitor (>=6000 tok/s bar)"
    status: planned
  - id: PMAT-716
    title: "MISSION: one pure-Rust binary that REPLACES + BEATS sklearn + PyTorch + Unsloth + Ollama/llama.cpp at what each does best"
    status: planned
  - id: PMAT-717
    title: "Pillar 1: BEAT scikit-learn at classical ML (pure Rust) — fit/predict faster, same accuracy, provably correct"
    status: planned
  - id: PMAT-718
    title: "Pillar 2: BEAT PyTorch at tensors + autograd + training (bounded task: train-to-loss faster / less memory)"
    status: planned
  - id: PMAT-719
    title: "Pillar 4: BEAT Ollama/llama.cpp at local quantized inference — 1.5x decode tok/s AND provable output-correctness"
    status: planned
  - id: PMAT-720
    title: "Add aprender::datasets module (load_iris, load_digits, load_california_housing via include_str! + make_classification/make_blobs/make_regression/make…"
    status: planned
  - id: PMAT-721
    title: "Falsify the LogisticRegression train_acc=0.505 observation: contract-gated test that LogReg reaches >=0.95 train-acc on a margin-separable 2-class se…"
    status: planned
  - id: PMAT-722
    title: "Commit the head-to-head sklearn-beat bench harness (apr vs sklearn) on digits classification + california regression, CSV/JSON out, CI-failing if apr…"
    status: planned
  - id: PMAT-723
    title: "Ship `apr qa --falsify` adversarial-corpus mode: bundle N>=5 deliberately-broken GGUF/APR artifacts (corrupt quant block, NaN weights, transposed lm_…"
    status: planned
  - id: PMAT-724
    title: "Add finite-difference gradient-correctness gate to core autograd as a proptest (perturb eps=1e-5, assert max|analytic-numeric|<1e-3 over matmul/relu/…"
    status: planned
  - id: PMAT-725
    title: "Build the CPU training-time beat-benchmark harness: fixed 2-layer MLP + fixed N=1024 synthetic regression, single-thread CPU, measure wall-clock-to-t…"
    status: planned
  - id: PMAT-726
    title: "PMAT-711: wire QLoRALayer (4-bit NF4 base) into the InstructPipeline CPU train loop (currently Vec<LoRALayer> f32 base, mod.rs:176); reuse the EXISTI…"
    status: planned
  - id: PMAT-727
    title: "PMAT-712: merge->GGUF as a contract-gated golden test — extend run_merge to optionally emit GGUF, add #[contract] obligation that merged-GGUF forward…"
    status: planned
  - id: PMAT-728
    title: "Run pinned PyTorch-CPU baseline (single OMP thread) on the IDENTICAL fixed MLP task; record wall-clock-to-target-loss + peak RSS via /usr/bin/time -v…"
    status: planned
  - id: PMAT-729
    title: "Unify classifier API onto Estimator (or add a Classifier trait with y:&[usize]) so cross_validate/grid_search work generically over LogReg/RF/GBM/SVM…"
    status: planned
  - id: PMAT-730
    title: "Add missing metrics roc_auc_score, log_loss, roc_curve, precision_recall_curve (VERIFIED ABSENT everywhere) + GradientBoostingRegressor to complete t…"
    status: completed
  - id: PMAT-731
    title: "PMAT-713: one-command `apr finetune --qlora --bits 4 --export gguf -o out.gguf` doing load->NF4-QLoRA->CPU-train->merge->export on a tiny fixture (no…"
    status: planned
  - id: PMAT-732
    title: "Wire apr-qa-differential-v1 ollama_parity into a real harness: `apr qa --differential --reference llama.cpp` reporting top-1 agreement >=95% + ppl-ga…"
    status: planned
  - id: PMAT-733
    title: "Add preprocessing encoders + sklearn-style Pipeline: OneHotEncoder, LabelEncoder, OrdinalEncoder, PolynomialFeatures, SimpleImputer, Normalizer + Pip…"
    status: completed
  - id: PMAT-734
    title: "PMAT-715: promote Unsloth from static const (score.rs:777) to a tracked competitor ROW in compute_training_step_scorecard reporting apr/unsloth tok/s…"
    status: planned
  - id: PMAT-735
    title: "Add non-linear SVM (RBF/poly kernel SVC) — sklearn SVC''s signature strength; current LinearSVM cannot separate non-linear data. VERIFIED ABSENT (no …"
    status: completed
  - id: PMAT-736
    title: "Publish a curated `apr registry` short-name catalog (`apr run qwen2.5-coder-1.5b` no URL) with each entry carrying its `apr qa` verdict as a trust ba…"
    status: planned
  - id: PMAT-737
    title: "[GPU-GATED — flag] CPU Q4_K decode parity: implement ggml-style pre-interleaved Q4_K weight layout at APR import (documented root cause, five-whys-16…"
    status: planned
  - id: PMAT-738
    title: "[GPU-GATED — NOT loop-buildable] Single-GPU QLoRA throughput run on lambda-vector to populate apr-vs-unsloth tok/s bar (6715.7/13659.7) — measurement…"
    status: planned
  - id: PMAT-739
    title: "[GPU-GATED — multi-week stretch] Drive GPU 1.5B Q4_K_M from 1.32x to 1.5x Ollama via FUSION-004 DP4A Q8_1 (only un-falsified lever; FUSION-003 alread…"
    status: planned
  - id: PMAT-740
    title: "ROADMAP GAP: add a classical-ML (scikit-learn) competitor category (CRUX-S-*) to the CRUX matrix"
    status: planned
  - id: PMAT-741
    title: "ROADMAP GAP: add a beat-benchmark contract kind (committed head-to-head, CI-failing, baseline pinned)"
    status: planned
  - id: PMAT-CLAUDE-PROXY-001
    title: "APR-CLAUDE-PROXY: `apr serve anthropic` — sovereign Messages-API drop-in backed by apr code + Qwen3-MoE"
    status: planned
  - id: PMAT-CODE-MCP-CLIENT-001
    title: "APR-CODE-MCP-CLIENT: Wire apr code to load external MCP servers via .mcp.json"
    status: planned
  - id: PMAT-CODE-PARITY-MATRIX-001
    title: "APR-CODE-PARITY: close the 20-feature gap vs Claude Code surfaced by the 2026-04-18 audit"
    status: planned
  - id: PMAT-MCP-PARITY-001
    title: "APR-MCP-PARITY: Expand aprender-mcp coverage from 9 → full 58-command apr surface"
    status: planned
  - id: PMAT-PERF-009-CUDA
    title: "New task: PMAT-PERF-009-CUDA"
    status: completed
  - id: PMAT-PERF-009-WIRE
    title: "New task: PMAT-PERF-009-WIRE"
    status: cancelled
  - id: PMAT-PERF-2X-TARGET
    title: "New task: PMAT-PERF-2X-TARGET"
    status: cancelled
  - id: PMAT-PERF-OPTIMIZE
    title: "New task: PMAT-PERF-OPTIMIZE"
    status: cancelled
  - id: "PMAT-PHASE6: Re-run Phase 6 (marathon + throughput) with self-converted GGUF"
    title: "New task: PMAT-PHASE6: Re-run Phase 6 (marathon + throughput) with self-converted GGUF"
    status: completed
  - id: PMAT-ROSETTA-001
    title: "New task: PMAT-ROSETTA-001"
    status: completed
  - id: PMAT-SHOWCASE-NEXT
    title: "New task: PMAT-SHOWCASE-NEXT"
    status: completed
  - id: "Phase 1: Classical Bayesian Inference - Conjugate Priors (Gamma-Poisson, Normal-InverseGamma, Dirichlet-Multinomial)"
    title: "New task: Phase 1: Classical Bayesian Inference - Conjugate Priors (Gamma-Poisson, Normal-InverseGamma, Dirichlet-Multinomial)"
    status: completed
  - id: "Phase 1: Graph Pathfinding Algorithms (Dijkstra, A*, APSP, shortest_path)"
    title: "New task: Phase 1: Graph Pathfinding Algorithms (Dijkstra, A*, APSP, shortest_path)"
    status: completed
  - id: "Phase 2: Bayesian Examples & Documentation - Create cargo examples and book chapters for Gamma-Poisson, Normal-InverseGamma, and Dirichlet-Multinomial conjugate priors"
    title: "New task: Phase 2: Bayesian Examples & Documentation - Create cargo examples and book chapters for Gamma-Poisson, Normal-InverseGamma, and Dirichlet-Multinomial conjugate priors"
    status: completed
  - id: "Phase 2: Components & Traversal (DFS, connected_components, SCCs, topological_sort)"
    title: "New task: Phase 2: Components & Traversal (DFS, connected_components, SCCs, topological_sort)"
    status: completed
  - id: "Phase 3: Community & Link Analysis (label_propagation, common_neighbors, adamic_adar)"
    title: "New task: Phase 3: Community & Link Analysis (label_propagation, common_neighbors, adamic_adar)"
    status: completed
  - id: "Phase 4: Graph Benchmarks & Spec Update"
    title: "New task: Phase 4: Graph Benchmarks & Spec Update"
    status: completed
  - id: "Push 4 commits to origin/main (Bayesian documentation and doctest fixes)"
    title: "New task: Push 4 commits to origin/main (Bayesian documentation and doctest fixes)"
    status: completed
  - id: "Push completed work to origin/main (5 commits)"
    title: "New task: Push completed work to origin/main (5 commits)"
    status: completed
  - id: ROADMAP-RECONCILE-2026-07-04
    title: "Reconciliation note (2026-07-04): status drift corrected + beats pivot recorded"
    status: completed
  - id: Round-49-falsification
    title: "New task: Round-49-falsification"
    status: cancelled
  - id: SHOWCASE-BRICK-001
    title: "New task: SHOWCASE-BRICK-001"
    status: completed
  - id: SVC-SMO-WSS-001
    title: "SVCRbf: upgrade simplified SMO to libsvm 2nd-order working-set selection (WSS)"
    status: planned
  - id: "Sovereign AI Integration Specification v1.1 - Complete architecture document with 10 peer-reviewed references, Toyota Way principles, and Rust safety improvements"
    title: "New task: Sovereign AI Integration Specification v1.1 - Complete architecture document with 10 peer-reviewed references, Toyota Way principles, and Rust safety improvements"
    status: completed
  - id: "Update roadmap status for completed TOP 10 algorithms (KNN, GBM, PCA)"
    title: "New task: Update roadmap status for completed TOP 10 algorithms (KNN, GBM, PCA)"
    status: completed
  - id: "Update trait signatures to use AprenderError instead of static str"
    title: "New task: Update trait signatures to use AprenderError instead of static str"
    status: completed
  - id: WAPR-CLI-STUBS
    title: "New task: WAPR-CLI-STUBS"
    status: cancelled
  - id: aprender-tsp-v0.1.0
    title: "release: aprender-tsp v0.1.0 - TSP solver with ACO, Tabu, GA, Hybrid and HuggingFace POC models"
    status: completed
  - id: "docs/specifications/qwen2.5-coder-showcase.md"
    title: "New task: docs/specifications/qwen2.5-coder-showcase.md"
    status: completed
  - id: nlp-text-preprocessing
    title: "New task: nlp-text-preprocessing"
    status: completed
  - id: qwen2.5-coder-showcase-demo
    title: "New task: qwen2.5-coder-showcase-demo"
    status: completed