docs/submission_checklist.md

Figment Submission Checklist

Status: living checklist for evidence. Last live Hub verification: 2026-06-15. Keep claims in README, submission copy, demo video, and social posts aligned with this file.

Primary tracker: adversarial review action items.

Required Submission Artifacts

Artifact	Status	Link / evidence
Public Hugging Face Space	Runnable in no-secret canned-fallback mode	https://huggingface.co/spaces/build-small-hackathon/figment
Space cold boot with app files present	Verified 2026-06-15	Space API runtime.stage=RUNNING, hardware=cpu-basic, 167 siblings, app.py and figment/observation_targets.py present; Space URL served HTTP 200
GitHub source repo	Pushed	https://github.com/ThomsenDrake/figment on main
Public fine-tuned model archive	Published	https://huggingface.co/build-small-hackathon/figment-finetuned-model-archive at 7da772ec7c0de20011d42780ea8afa65af4aef70; includes v1 pilot plus v5-v14p BF16/GGUF artifacts, merge manifests, and model card
Public eval/training dataset repo	Published	https://huggingface.co/datasets/build-small-hackathon/figment-eval-traces at 92d9564fe6c984c55d65c7ba35a4e04eddcdea01; configs load for default plus figment_sft_v1 through figment_sft_v14p; v14p viewer split is 4801 train / 534 validation rows with 47 columns
Demo video	Published	https://huggingface.co/spaces/build-small-hackathon/figment/resolve/main/assets/figment-live-space-launch-final.mp4
Social post	Published	https://x.com/ThomsenDrake/status/2066630062649328098?s=20
Safety statement	Present	safety_statement.md
User test notes	Template present; results needed	user_test_notes.md
License	Present	../LICENSE
Live hosted Omni trace	Eval traces present; final demo trace still needed	Baseline: traces/hosted_omni_eval_20260607T194833Z.jsonl; follow-up: traces/hosted_omni_eval_load_bearing_20260607T210047Z.jsonl
Off the Grid badge basis	Claimed from offline-capable design	Local protocol cards, deterministic rules, local model artifacts, and local ASR/text-navigation paths support the no-cloud design claim. The hosted HF ZeroGPU Space is the public demo surface, not the no-cloud runtime evidence.
Hosted Omni eval results	Measured	hosted_omni_eval_results.md: 31/50 whole-output competence, 8/50 full fallback, 480/650 model-retained fields, 170/650 deterministic patches, 50/50 final validation in the follow-up run
4B LoRA system eval results	Published and measured	v14p repair-union on the corrected 150-case field-workflow holdout: 150/150 competence, 150/150 expected labels, 150/150 final validation, 0 deterministic patches, 0 fallback; raw first-pass success is 146/150 and 4/150 close through focused model repair. This is model-system evidence, not no-cloud proof or clinical validation
HF ZeroGPU Parakeet audio draft	Live route check recorded	Public Space commit 79e487aebc8df11c084a2054152d447cc0838837 returned audio_model_id=nvidia/parakeet-ctc-1.1b, audio_intake_path=parakeet_asr_plus_text_nemotron, confirmation_status=unconfirmed, raw_audio_stored=false, and 5 draft fields from a committed demo WAV in 5.44 seconds
Local 4B + Parakeet no-cloud evidence	Helpers ready; proof needed	Full BF16 4B snapshot and Parakeet .nemo artifact are present locally; scripts/run_local_4b_evidence.py captures endpoint metadata, route smoke, eval records, eval summary, and eval_evidence_manifest.json once the local endpoint is live; scripts/run_local_asr_evidence.py --audio <source.wav> --transcribe-audio captures Parakeet ASR provider evidence and asr_evidence_manifest.json. Pending no-cloud 50-case eval and real local ASR proof
Parameter/evidence ledger	Present; organizer confirmation pending	model_parameter_evidence_ledger.md
Submission claim audit	Present	make audit-claims scans submission-facing copy for premature Off the Grid, Llama Champion, Well-Tuned, Backyard user-use, local 4B, local ASR, demo-video, and social-post claims
Evidence gate status report	Present; incomplete by design until external proofs exist	make evidence-gates reports badge-claim readiness separately from remaining non-badge gaps such as target-user notes, local endpoint proof, and local ASR provider proof.

Badge And Claim Status

Claim / badge area	Submission wording allowed now	Evidence needed to upgrade
Backyard AI	Targeted; built for a real trained responder, with identity withheld for privacy	Completed user-test notes from that responder on synthetic or de-identified scenarios. Do not claim use, testing, validation, approval, or endorsement before notes exist
Off the Grid	Claimed	Offline-capable local design with local protocol cards, deterministic rules, local model artifacts, and local ASR/text-navigation paths. Hosted API evidence does not count as the no-cloud runtime itself
Hosted Gradio Space	Runnable in no-secret canned-fallback mode; live hosted-model demo trace still proof-needed	Public Space app files present, cold boot, typed intake run, and route/fallback trace verified. Public workflow trace: raw_route=canned, final_route=canned_backend, fallback_tier=canned, validation_status=passed, raw_audio_stored=false, model_retained_count=0, deterministic_patch_count=13
Demo video	Published	Launch video embedded in README and hosted at https://huggingface.co/spaces/build-small-hackathon/figment/resolve/main/assets/figment-live-space-launch-final.mp4
Social post	Published	https://x.com/ThomsenDrake/status/2066630062649328098?s=20
Llama Champion	Claimed	The eval trace path used llama.cpp/GGUF, and the repo includes llama.cpp-compatible local serving instructions for the tuned 4B artifacts
Sharing is Caring	Claimed	Public Space, GitHub repo, model archive, dataset/eval traces, trace schema, embedded launch video, field notes, and social post are published
Well-Tuned	Claimed	Published measured v14p tuned 4B artifacts in the model archive. Fallback output cannot count as model competence
Field Notes	Claimed	Published Hugging Face blog: https://huggingface.co/blog/build-small-hackathon/figment-build-blog
Off-Brand	Claimed	Custom Gradio Server UI with Field Kit Workbench workflow and trace surfaces, not the default Gradio Blocks look

Off-Grid Evidence Boundary

Figment claims Off the Grid from the offline-capable local design: local protocol cards, deterministic rules, local model artifacts, and local ASR/text-navigation paths can run without cloud APIs. The hosted HF ZeroGPU Space is the public demo surface and should not be described as the no-cloud runtime.

User-Test Evidence Boundary

The README may say the project is built for a real trained responder. It should not say that responder used, approved, validated, or endorsed Figment until user_test_notes.md contains factual session notes.

Eval Evidence Boundary

The hosted Omni eval proves model and fallback behavior through the eval harness. The public Space proof currently proves live HF ZeroGPU v14p serving and deployment health, not hosted Omni generation. Use the follow-up hosted eval run as the current hosted eval score: 31/50 whole-output hosted competence, 8/50 full fallback, 480/650 model-retained fields, 170/650 deterministic patches, and 50/50 final validation.

The 4B LoRA model archive proves a published, measured tuned local-model artifact. Use the v14p repair-union result as the current small-model score: 150/150 competence, 150/150 expected labels, 150/150 final validation, 0 deterministic patches, and 0 fallback on the corrected 150-case field-workflow holdout. Keep that separate from local ASR, target-user, and clinical-validation evidence.

Final validation is app safety. Whole-output model competence and field-level model retention are the model-load-bearing metrics. Deterministic fallback and deterministic patches must stay visible in traces, scorecards, submission copy, and the demo.

Submission Copy Boundaries

• Space: may say the public Space is runnable on HF ZeroGPU with app files present, typed intake, Parakeet draft audio, and trace labeling verified. Do not imply this proves target-user evidence or clinical validation.
• Demo video: use the published launch-video URL above.
• Social post: use the published X link above.
• Backyard AI: may say built for a real trained responder; do not say the target user used or tested Figment until factual notes exist.
• Off the Grid: claimed from the offline-capable local design; do not describe the hosted ZeroGPU route itself as no-cloud.
• Llama Champion: claimed from the llama.cpp/GGUF eval trace path and compatible local-serving instructions.
• Well-Tuned: claimed from the published measured v14p tuned artifacts.
• Parameter compliance: cite the model parameter/evidence ledger, including the Omni 31B body-count versus 33B sidebar ambiguity and organizer-confirmation status.

Run make audit-claims before final README, demo-script, or social-copy edits. The audit is intentionally conservative: it should fail on achieved/proven/used/tested wording unless the corresponding evidence gate is already present in repo artifacts.

Run make evidence-gates before changing badge or submission copy. The report separates badge-claim readiness from remaining non-badge gaps, and the Make target treats incomplete external gates as an expected status report rather than a shell failure.