ParetoOptimal/repro-evomas
0
1{2 "schema_version": 1,3 "title": "Repro - Evolutionary Generation of Multi-Agent Systems",4 "emoji": "๐ฏ",5 "space_id": "ParetoOptimal/repro-evomas",6 "paper": {7 "arxiv_id": "2602.06511"8 },9 "tags": [10 "icml2026-repro",11 "paper-ic0AGRIkmY"12 ],13 "updated_at": "2026-07-16T23:50:55+00:00",14 "root": {15 "slug": "index",16 "title": "Repro - Evolutionary Generation of Multi-Agent Systems",17 "file": "pages/index.md",18 "children": [19 {20 "slug": "paper-approach",21 "title": "Paper & approach",22 "file": "pages/paper-approach/page.md",23 "children": []24 },25 {26 "slug": "claim-2-bbeh-accuracy-vs-baselines-toy",27 "title": "Official claim 1: +10.5 over EvoAgent on BBEH, +7.1 on WorkBench (toy-scale direction test)",28 "file": "pages/claim-2-bbeh-accuracy-vs-baselines-toy/page.md",29 "children": []30 },31 {32 "slug": "claim-3-swe-bench-verified-79-1-audit-only",33 "title": "Official claim 2: 79.1% on SWE-Bench-Verified with Claude-4.5-Sonnet (audit only)",34 "file": "pages/claim-3-swe-bench-verified-79-1-audit-only/page.md",35 "children": []36 },37 {38 "slug": "claim-5-execution-rate",39 "title": "Official claim 3: higher executability and runtime robustness (execution-rate measurements)",40 "file": "pages/claim-5-execution-rate/page.md",41 "children": []42 },43 {44 "slug": "claim-1-structured-configuration-generation",45 "title": "Evidence A: the EvoMAS mechanism โ structured configs + evolutionary loop (underpins official claims 1 and 3)",46 "file": "pages/claim-1-structured-configuration-generation/page.md",47 "children": []48 },49 {50 "slug": "claim-4-meta-model-sensitivity-toy",51 "title": "Evidence B: meta-model sensitivity (supports official claims 1 and 3)",52 "file": "pages/claim-4-meta-model-sensitivity-toy/page.md",53 "children": []54 },55 {56 "slug": "evomas-repro",57 "title": "evomas-repro",58 "file": "pages/evomas-repro/page.md",59 "children": []60 },61 {62 "slug": "conclusion",63 "title": "Conclusion",64 "file": "pages/conclusion/page.md",65 "children": []66 }67 ]68 },69 "agent_view_tokens": 11860,70 "revision": "1784245855730112194"71}