Training-Harness (ACO, Multi-Fidelity), Prüfstand-Benchmark, Agenten-README

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
team3
2026-07-04 12:47:19 +02:00
parent 8d8f6c8e51
commit 8488737303
14 changed files with 772 additions and 267 deletions

View File

@@ -8,43 +8,43 @@ bewusst NICHT hier — die Messlatte darf nie Teil des Suchraums sein.
PARAMS: dict[str, dict] = {
# Recherche / Inventar
"RESEARCH_THEMA_AGENTS": {"default": 5, "min": 2, "max": 8, "step": 1, "kategorie": "qualitaet"},
"RESEARCH_READERS": {"default": 2, "min": 1, "max": 3, "step": 1, "kategorie": "qualitaet"},
"RESEARCH_SECTION_CHARS": {"default": 12000, "min": 6000, "max": 24000, "step": 3000, "kategorie": "tokens"},
"DEDUP_PAIR_FLOOR": {"default": 0.6, "min": 0.45, "max": 0.8, "step": 0.05, "kategorie": "auswahl"},
"DEDUP_TITLE_AUTO": {"default": 0.95, "min": 0.9, "max": 0.99, "step": 0.01, "kategorie": "auswahl"},
"DEDUP_GLOBAL_FLOOR": {"default": 0.65, "min": 0.5, "max": 0.8, "step": 0.05, "kategorie": "auswahl"},
"DEDUP_PAIRS_CHUNK": {"default": 40, "min": 15, "max": 80, "step": 10, "kategorie": "laufzeit"},
"FILTER_CHUNK": {"default": 35, "min": 15, "max": 60, "step": 10, "kategorie": "laufzeit"},
"FILTER_RECHECK_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet"},
"CONSOLIDATION_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet"},
"RESEARCH_THEMA_AGENTS": {"default": 5, "min": 2, "max": 8, "step": 1, "kategorie": "qualitaet", "fidelity": "voll"},
"RESEARCH_READERS": {"default": 2, "min": 1, "max": 3, "step": 1, "kategorie": "qualitaet", "fidelity": "voll"},
"RESEARCH_SECTION_CHARS": {"default": 12000, "min": 6000, "max": 24000, "step": 3000, "kategorie": "tokens", "fidelity": "voll"},
"DEDUP_PAIR_FLOOR": {"default": 0.6, "min": 0.45, "max": 0.8, "step": 0.05, "kategorie": "auswahl", "fidelity": "voll"},
"DEDUP_TITLE_AUTO": {"default": 0.95, "min": 0.9, "max": 0.99, "step": 0.01, "kategorie": "auswahl", "fidelity": "voll"},
"DEDUP_GLOBAL_FLOOR": {"default": 0.65, "min": 0.5, "max": 0.8, "step": 0.05, "kategorie": "auswahl", "fidelity": "voll"},
"DEDUP_PAIRS_CHUNK": {"default": 40, "min": 15, "max": 80, "step": 10, "kategorie": "laufzeit", "fidelity": "voll"},
"FILTER_CHUNK": {"default": 35, "min": 15, "max": 60, "step": 10, "kategorie": "laufzeit", "fidelity": "voll"},
"FILTER_RECHECK_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet", "fidelity": "voll"},
"CONSOLIDATION_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet", "fidelity": "voll"},
# Subbausteine
"SUBBLOCK_CHUNK": {"default": 10, "min": 4, "max": 20, "step": 2, "kategorie": "laufzeit"},
"SUBBLOCK_MIN": {"default": 5, "min": 2, "max": 10, "step": 1, "kategorie": "auswahl"},
"SUBBLOCK_MAX_ROUNDS": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "auswahl"},
"SUBBLOCK_EXTRA_ROUNDS": {"default": 2, "min": 0, "max": 4, "step": 1, "kategorie": "auswahl"},
"SUBBLOCK_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet"},
"SUBBLOCK_CHUNK": {"default": 10, "min": 4, "max": 20, "step": 2, "kategorie": "laufzeit", "fidelity": "board2"},
"SUBBLOCK_MIN": {"default": 5, "min": 2, "max": 10, "step": 1, "kategorie": "auswahl", "fidelity": "board2"},
"SUBBLOCK_MAX_ROUNDS": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "auswahl", "fidelity": "board2"},
"SUBBLOCK_EXTRA_ROUNDS": {"default": 2, "min": 0, "max": 4, "step": 1, "kategorie": "auswahl", "fidelity": "board2"},
"SUBBLOCK_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet", "fidelity": "board2"},
# Facts / Artefakte / Fragen
"FACTS_CHUNK_SUBS": {"default": 10, "min": 4, "max": 20, "step": 2, "kategorie": "laufzeit"},
"FACTS_CHECK_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet"},
"QUESTION_CHUNK_SUBS": {"default": 25, "min": 10, "max": 50, "step": 5, "kategorie": "laufzeit"},
"QUESTION_MAX_ROUNDS": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet"},
"ARTEFACT_CHUNK_SUBS": {"default": 25, "min": 10, "max": 50, "step": 5, "kategorie": "laufzeit"},
"FACTS_CHUNK_SUBS": {"default": 10, "min": 4, "max": 20, "step": 2, "kategorie": "laufzeit", "fidelity": "board2"},
"FACTS_CHECK_PANEL": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet", "fidelity": "board2"},
"QUESTION_CHUNK_SUBS": {"default": 25, "min": 10, "max": 50, "step": 5, "kategorie": "laufzeit", "fidelity": "board2"},
"QUESTION_MAX_ROUNDS": {"default": 3, "min": 1, "max": 5, "step": 1, "kategorie": "qualitaet", "fidelity": "board2"},
"ARTEFACT_CHUNK_SUBS": {"default": 25, "min": 10, "max": 50, "step": 5, "kategorie": "laufzeit", "fidelity": "board2"},
# Embedding-Schwellen (Auswahl-Kern)
"SUB_VARIANT_COS": {"default": 0.90, "min": 0.85, "max": 0.96, "step": 0.01, "kategorie": "auswahl"},
"SEED_COVER_COS": {"default": 0.80, "min": 0.7, "max": 0.9, "step": 0.02, "kategorie": "auswahl"},
"SUB_DUP_KANDIDAT_COS": {"default": 0.75, "min": 0.65, "max": 0.85, "step": 0.02, "kategorie": "auswahl"},
"EMBEDDING_BLOCK_FLOOR": {"default": 0.5, "min": 0.35, "max": 0.65, "step": 0.05, "kategorie": "auswahl"},
"CROSS_CHUNK_PAARE": {"default": 40, "min": 15, "max": 80, "step": 10, "kategorie": "laufzeit"},
"SUB_VARIANT_COS": {"default": 0.90, "min": 0.85, "max": 0.96, "step": 0.01, "kategorie": "auswahl", "fidelity": "board2"},
"SEED_COVER_COS": {"default": 0.80, "min": 0.7, "max": 0.9, "step": 0.02, "kategorie": "auswahl", "fidelity": "board2"},
"SUB_DUP_KANDIDAT_COS": {"default": 0.75, "min": 0.65, "max": 0.85, "step": 0.02, "kategorie": "auswahl", "fidelity": "board2"},
"EMBEDDING_BLOCK_FLOOR": {"default": 0.5, "min": 0.35, "max": 0.65, "step": 0.05, "kategorie": "auswahl", "fidelity": "voll"},
"CROSS_CHUNK_PAARE": {"default": 40, "min": 15, "max": 80, "step": 10, "kategorie": "laufzeit", "fidelity": "board2"},
# Guide
"MAX_WRITER_ROUNDS": {"default": 2, "min": 1, "max": 3, "step": 1, "kategorie": "qualitaet"},
"GATE_FIX_MIN": {"default": 3, "min": 1, "max": 6, "step": 1, "kategorie": "qualitaet"},
"WRITER_SPLIT_SUBS": {"default": 30, "min": 15, "max": 45, "step": 5, "kategorie": "qualitaet"},
"MAX_WRITER_ROUNDS": {"default": 2, "min": 1, "max": 3, "step": 1, "kategorie": "qualitaet", "fidelity": "board2"},
"GATE_FIX_MIN": {"default": 3, "min": 1, "max": 6, "step": 1, "kategorie": "qualitaet", "fidelity": "board2"},
"WRITER_SPLIT_SUBS": {"default": 30, "min": 15, "max": 45, "step": 5, "kategorie": "qualitaet", "fidelity": "board2"},
# Engine / Kosten
"CONSENSUS_GRACE": {"default": 300, "min": 0, "max": 600, "step": 60, "kategorie": "laufzeit"},
"MAX_RESTARTS": {"default": 2, "min": 1, "max": 3, "step": 1, "kategorie": "laufzeit"},
"EVIDENCE_BUDGET_CHARS": {"default": 48000, "min": 16000, "max": 64000, "step": 8000, "kategorie": "tokens"},
"QUELLE_RELEVANZ_CHUNK": {"default": 12, "min": 6, "max": 24, "step": 3, "kategorie": "laufzeit"},
"CONSENSUS_GRACE": {"default": 300, "min": 0, "max": 600, "step": 60, "kategorie": "laufzeit", "fidelity": "board2"},
"MAX_RESTARTS": {"default": 2, "min": 1, "max": 3, "step": 1, "kategorie": "laufzeit", "fidelity": "board2"},
"EVIDENCE_BUDGET_CHARS": {"default": 48000, "min": 16000, "max": 64000, "step": 8000, "kategorie": "tokens", "fidelity": "board2"},
"QUELLE_RELEVANZ_CHUNK": {"default": 12, "min": 6, "max": 24, "step": 3, "kategorie": "laufzeit", "fidelity": "voll"},
}