κ
Kytos Observatory

Day 31 of 78 · Run #26 · 2026-09-19

H1 rank-64 transfer + dual-moment counts on context C only — overall +0.042 (rank 546), regression vs k011; the k017 offline win did not transfer to the leaderboard

Overall score 0.042 · Perturbation discrimination (pds) 0.280 · +5 more

Strategyh1-rank64-transfer (context C) + dual-moment counts; A/B on k011 champion
What’s next
Audit & metrics Overall score 0.042 · Perturbation discrimination (pds) 0.280 · +5 more + −

No audit flags.

Differential expression volcano plot (log2FC vs -log10 p)
Pre-registered hypotheses (1)
  • a learned H1 signature improves context-C predictions over the k011 champion
Evidence Literature & entity enrichment + −

Literature

0 genes

Tavily · auxiliary, not scored

Literature pending — run tools/enrich_literature.py.

Field context

VCC / perturbation research

Field research pending — run tools/enrich_newsroom.py.

Biomedical NER

0 entities

Pioneer GLiNER2 · fine-tuned on Tavily literature when available · regex fallback offline

NER pending — run tools/pioneer_ner.py --train once, then tools/pioneer_ner.py --run experiments/<run-id>.

Narrative digest run digest · traces to facts.json + −

Narrative pending — run tools/render_narrative.py.

Trust & provenance Self-tests + reproduce command + −
🤖
Dr. Kytos Biological Observer Pipeline Execution Trace

6 pipeline steps · sourced from committed artifacts only

○
cell_eval.score_perturbations()

No metrics CSVs found — run cell-eval and commit results.

✓
kytos.audit.evaluate_rules()

Ran 5 deterministic rules · 0 flags raised (none fired).

○
tavily.search_literature()

Literature enrichment not yet run — execute tools/enrich_literature.py.

○
pioneer.ner_extract_entities()

NER extraction not yet run — execute tools/pioneer_ner.py.

○
narrative.synthesize_digest()

Narrative not yet generated — execute tools/render_narrative.py.

○
verification.self_audit()

Verification artifacts not yet committed — run planted_signal.py, holo_audit.py, check_narrative.py.

PENDINGPlanted-signal self-testnot run yet

Run python tools/planted_signal.py --json …/verification/planted_signal.json.

PENDINGIndependent agent auditnot run yet

Run python tools/holo_audit.py --run experiments/<run-id>.

PENDINGNarrative grounding checknot run yet

Run python tools/check_narrative.py --run experiments/<run-id> after the digest is generated.