binzhango's picture
Update logbook: Repro - ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering
14ea6d3 verified
Raw
History Blame Contribute Delete
2.4 kB
{
"schema_version": 1,
"title": "Repro - ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering",
"emoji": "🧪",
"space_id": "binzhango/repro-ml-agent-autonomous-ml-engineering-v2",
"paper": {
"arxiv_id": "2505.23723"
},
"tags": [
"icml2026-repro",
"paper-kcPPWaoegr",
"independent-execution"
],
"updated_at": "2026-07-18T01:59:27+00:00",
"root": {
"slug": "index",
"title": "Repro - ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering",
"file": "pages/index.md",
"children": [
{
"slug": "reproduction-contract-and-release-audit",
"title": "Reproduction Contract and Release Audit",
"file": "pages/reproduction-contract-and-release-audit/page.md",
"children": []
},
{
"slug": "claim-1-7b-ml-agent-vs-larger-agents",
"title": "Claim 1: 7B ML-Agent vs Larger Agents",
"file": "pages/claim-1-7b-ml-agent-vs-larger-agents/page.md",
"children": []
},
{
"slug": "claim-2-nine-task-training-and-held-out-generalization",
"title": "Claim 2: Nine-Task Training and Held-Out Generalization",
"file": "pages/claim-2-nine-task-training-and-held-out-generalization/page.md",
"children": []
},
{
"slug": "claim-3-cifar-10-accuracy",
"title": "Claim 3: CIFAR-10 Accuracy",
"file": "pages/claim-3-cifar-10-accuracy/page.md",
"children": []
},
{
"slug": "claim-4-behaviorally-distant-exploration",
"title": "Claim 4: Behaviorally Distant Exploration",
"file": "pages/claim-4-behaviorally-distant-exploration/page.md",
"children": []
},
{
"slug": "claim-5-ml-specific-reward-branches",
"title": "Claim 5: ML-Specific Reward Branches",
"file": "pages/claim-5-ml-specific-reward-branches/page.md",
"children": []
},
{
"slug": "conclusion",
"title": "Conclusion",
"file": "pages/conclusion/page.md",
"children": []
},
{
"slug": "ml-agent-repro-17897-v2-artifacts",
"title": "ml-agent-repro-17897-v2-artifacts",
"file": "pages/ml-agent-repro-17897-v2-artifacts/page.md",
"children": []
}
]
},
"agent_view_tokens": 7331,
"revision": "1784339967578907000"
}