PUMA Vault

Etiqueta: methodology

91 artículos con esta etiqueta.

15 jun 2026
Extreme Programming Explained: Embrace Change
15 jun 2026
Work the System: The Simple Mechanics of Making More and Working Less
15 jun 2026
The E-Myth Revisited: Why Most Small Businesses Don't Work and What to Do About It
15 jun 2026
The Goal: A Process of Ongoing Improvement
15 jun 2026
Scaling Up: How a Few Companies Make It... and Why the Rest Don't
15 jun 2026
The Frictionless Organization: Deliver Great Customer Experiences with Less Effort
15 jun 2026
Pensar con Prompts: La guía definitiva de la ingeniería de prompts
15 jun 2026
Traction: Get a Grip on Your Business
15 jun 2026
PN — The project asks whether local language models can manage technology…
15 jun 2026
PN — PUMA is a recursive acronym ('PUMA Understanding & Management w Agents')
15 jun 2026
PN — A historical precedent (an early automated content bot that propagated a…
15 jun 2026
PN — Technology project management in multi-project organizations suffers a…
15 jun 2026
PN — Problem one
15 jun 2026
PN — Problem two
15 jun 2026
PN — Problem three
15 jun 2026
PN — Reproducibility gap: in the most complete field review (85 papers across…
15 jun 2026
PN — Gap two
15 jun 2026
PN — Gap three
15 jun 2026
PN — PUMA occupies a deliberately narrow niche
15 jun 2026
PN — Meeting the estimation MVP threshold and the H2 hypothesis contrast are…
15 jun 2026
PN — A transferable lesson: on a long-tailed story-point distribution, the…
15 jun 2026
PN — The title question is answered cautiously
15 jun 2026
PN — The H2 non-rejection is reframed as a valuable methodological finding…
15 jun 2026
PN — The environmental framing
15 jun 2026
PN — The diversity/inclusion stance is open-source MIT release executable…
15 jun 2026
PN — The triage task uses a public dataset of 50,000+ real issues from…
15 jun 2026
PN — The estimation task uses a public dataset of 31,294 real user stories with…
15 jun 2026
PN — The main research question asks whether there are statistically…
15 jun 2026
PN — Hypotheses follow Popper's demarcation criterion: each is a refutable null…
15 jun 2026
PN — H1 (triage): the null states no model/strategy configuration beats the…
15 jun 2026
PN — H2 (estimation): the null states no example-enriched configuration beats…
15 jun 2026
PN — The work adopts Design Science Research, operationalized through six…
15 jun 2026
PN — Inference runs locally via a model server exposing an OpenAI-compatible…
15 jun 2026
PN — The MVP's canonical model is a 3B-parameter quantized model (~1.9 GB),…
15 jun 2026
PN — The experiment is a reduced factorial design with two main independent…
15 jun 2026
PN — Estimation maps user stories to a canonical Fibonacci story-point value…
15 jun 2026
PN — Reproducibility uses five mechanisms: fixed seed=42 propagated everywhere,…
15 jun 2026
PN — Four ethical risks are explicitly managed: role displacement…
15 jun 2026
Keshav Three-Pass and MIT AI Lab Three Questions: An Integrated Academic Reading Protocol for PUMA
15 jun 2026
PARA, GTD, and Zettelkasten work as complementary engines, not competing systems
15 jun 2026
Marco Veritas — AI Transparency and Academic Integrity Framework
15 jun 2026
PN — The system is a six-layer modular architecture (provisioning/hardware…
15 jun 2026
PN — Seven prompting-strategy families are implemented as reusable templates…
15 jun 2026
PN — Contextual-anchoring prepends operational definitions of each class/level…
15 jun 2026
PN — Triage maps free-text issues to one of four priority labels via an…
15 jun 2026
PN — Carbon footprint is measured with an offline emissions tracker activated…
15 jun 2026
PN — The literature review combines a structured three-pass reading method with…
15 jun 2026
PN — The artifact is built with Spec-Driven Development, where versioned…
15 jun 2026
PN — Statistical analysis uses the Wilcoxon signed-rank test (non-parametric,…
15 jun 2026
PN — Open-science stance
15 jun 2026
RES — Nine candidate open-weights models were selected (1B-12B across four families)
15 jun 2026
RES — The canonical triage baseline (3B + contextual-anchoring, N=200) achieves…
15 jun 2026
RES — Per-class triage: Critical is best discriminated (F1=0.816) via explicit…
15 jun 2026
RES — The triage confusion matrix shows errors confined to adjacent severities…
15 jun 2026
RES — In the triage sweep (N=100) the largest 12B model leads on F1-macro…
15 jun 2026
RES — Model quality is non-monotonic with size within families
15 jun 2026
RES — A keyword-heuristic baseline originally planned for triage was…
15 jun 2026
RES — The best estimation configuration achieves MAE=1.86 SP (7B model,…
15 jun 2026
RES — H1 is REJECTED: the canonical triage config beats the majority-class…
15 jun 2026
RES — H2 is NOT REJECTED: the best estimation config (7B, MAE=1.86 SP) does not…
15 jun 2026
RES — The full sweep covers 27 conditions (9 models x 3 scenarios x 100 = 2,700…
15 jun 2026
RES — A single non-parseable model family (parse-failure rate >= 0.98) consumed…
15 jun 2026
RES — Small models lead in raw efficiency (quality per gram CO2) for triage and…
15 jun 2026
RES — Estimated against frontier commercial models, the local 3B model's…
15 jun 2026
RES — Latency on the reference laptop profile is consistently under 3 s (P95)…
15 jun 2026
RES — All five MVP acceptance criteria are met comfortably (F1-macro, MAE,…
15 jun 2026
RES — Three consecutive warm-state runs of the canonical config produce…
15 jun 2026
RES — Qualitative triage error analysis codes four recurring types
15 jun 2026
RES — Contextual-anchoring is identified as an effective prompt-engineering…
15 jun 2026
RES — Native integration of sustainability as a first-class experimental…
15 jun 2026
RES — The empirical answer to the research question is asymmetric
15 jun 2026
RES — Bit-exact warm-state reproducibility for non-trivial local-LLM tasks is a…
15 jun 2026
RES — Reproducible evaluation on consumer laptop hardware requires attention to…
15 jun 2026
RES — Several methodological findings during validation (script renaming,…
15 jun 2026
RES — A cold-vs-warm drift of up to 0.006 F1-macro points was observed between…
15 jun 2026
DEC — The work covers three interdependent PMO tasks (triage, estimation,…
15 jun 2026
DEC — Scope is five stages of increasing complexity: stages 1-2 (triage,…
15 jun 2026
DEC — MVP acceptance is quantitative, verified at end of stage 2
15 jun 2026
DEC — The canonical baseline uses N=200 (deep validation of one config for tight…
15 jun 2026
DEC — A known limitation is imprecise GPU energy attribution via the local…
15 jun 2026
DEC — The implementation philosophy is 'harness engineering': effort…
15 jun 2026
DEC — Optimal model choice is task-specific, invalidating single-model…
15 jun 2026
DEC — A token-based economic valuation complements hour-based costing for…
15 jun 2026
DEC — Future work prioritizes inferential closure of H1/H2 against competitive…
15 jun 2026
DEC — Future work includes multi-seed validation (3-5 additional seeds) to…
15 jun 2026
DEC — Future work includes family-specific output parsers to support reasoning…
15 jun 2026
PRISMA Protocol — Systematic Literature Review for PUMA
15 jun 2026
PUMA Vault — Complete User Guide
01 may 2026
🧠 MOC — Methods & Frameworks
01 mar 2026
EGI — Exploración Guiada Interactiva (Guided Interactive Exploration)
01 mar 2026
RCOIF — Role·Context·Objective·Instructions·Format

Creado con Quartz v4.5.2 © 2026

GitHub
Discord Community