perplexity
55d53c513c
feat: Phase 3.5 — DPO training pair generation from Deep Dive pipeline
...
CI / test (pull_request) Failing after 22s
CI / validate (pull_request) Failing after 15s
Review Approval Gate / verify-review (pull_request) Failing after 2s
Wire arXiv relevance filter output directly into DPO pair generation,
closing the loop between research synthesis and overnight training data.
New module: dpo_generator.py
- DPOPairGenerator class with 3 pair strategies:
* summarize: paper → fleet-grounded analysis (chosen) vs generic (rejected)
* relevance: 'what matters to Hermes?' → scored context vs vague
* implication: 'what should we do?' → actionable insight vs platitude
- Extracts synthesis excerpts matched to each ranked item
- Outputs to ~/.timmy/training-data/dpo-pairs/deepdive_{timestamp}.jsonl
- Format: {prompt, chosen, rejected, task_type, evidence_ids,
source_session, safety_flags, metadata}
Pipeline changes (pipeline.py):
- Import DPOPairGenerator with graceful degradation
- Initialize from config deepdive.training.dpo section
- Execute as Phase 3.5 between synthesis and audio
- DPO results included in pipeline return dict
- Wrapped in try/except — DPO failure never blocks delivery
Config changes (config.yaml):
- New deepdive.training.dpo section with:
enabled, output_dir, min_score, max_pairs_per_run, pair_types
Integration tested: 2 mock items × 3 pair types = 6 valid JSONL pairs.
Chosen responses consistently richer than rejected (assert-verified).
2026-04-13 02:24:04 +00:00
34862cf5e5
feat(fleet): promote Ollama to first-class provider, assign Gemma 4 across fleet
...
Deploy Nexus / deploy (push) Failing after 3s
Staging Verification Gate / verify-staging (push) Failing after 3s
- lazarus-registry.yaml: replace big_brain/RunPod with local ollama/gemma4:12b
- fleet-routing.json: assign ollama:gemma4:12b to carnice, bilbobagginshire, substratum
- intelligence/deepdive/config.yaml: local model -> gemma4:12b
2026-04-07 15:55:52 +00:00
