bundle/v0/paper.json

{
  "schema_version": 1,
  "slug": "loss-aware-dro-ot",
  "title": "Loss-Aware Distributionally Robust Optimization via Trainable Optimal Transport Ambiguity Sets",
  "submission": "12512",
  "openreview": "K1EPPO9t2c",
  "arxiv": "2509.12689v1",
  "venue": "ICML 2026",
  "presentation": "Spotlight",
  "state": "PREPARATION_HOLD",
  "admission_decision": "ADMIT",
  "target_scale": "paper",
  "current_evidence_scale": "scientific_canary",
  "preparation_scaffold": {
    "paper_scale_task_count": 14000,
    "deterministic_seed_schedule": true,
    "raw_artifact_contract": true,
    "failure_ledger_contract": true,
    "claim_evaluator": true,
    "scientific_method_implemented": true,
    "scientific_method_scope": "gaussian_portfolio_empirical_portfolio_absolute_regression_squared_regression",
    "local_scientific_canary_passed": true,
    "presentation_blocked_until_evidence": true
  },
  "scored_claim_count": 6,
  "active_claim_source": "claims_anchored.json",
  "active_claims_lock": "configs/official_claims_lock.json",
  "scientific_claims": {"A1": "HOLD", "A2": "HOLD", "A3": "HOLD", "A4": "HOLD", "A5": "HOLD", "A6": "HOLD"},
  "official_source": {
    "url": "https://github.com/JonasOhn/trainable-ot-dro",
    "default_branch": "main",
    "revision": "6e8d18f023f4f2e920af22dc1ef32b292dbb9631",
    "commit_date": "2026-05-07T12:27:43Z",
    "tree": "0e118a88c77f20f09b8748de6810676eb39176d3",
    "license_declared": "MIT",
    "license_file_present": false,
    "license_declaration_path": "pyproject.toml",
    "license_declaration_sha256": "b441577fa99f44816083dc6c21696ad702e4c59a8ab8de49b763c9df92f58661"
  },
  "paper_source": {
    "url": "https://arxiv.org/abs/2509.12689",
    "version": "v1",
    "pdf_sha256": "761ac2afdaf4d0b87fcf89299cb7639b4edcc9137466aa27b4b15ee63707e41d"
  },
  "openresearch": {
    "required": true,
    "successful_baseline": false,
    "successful_child": false
  },
  "review": {
    "kimi_k3": false,
    "claude_fable_5_xhigh": false,
    "gpt_5_6_sol_xhigh": false,
    "challenge_validator": false,
    "user_approved": false
  },
  "authority": {
    "source_acquisition": false,
    "environment_build": false,
    "scientific_launch": false,
    "remote_compute": false,
    "paid_service": false,
    "ready_transition": false,
    "publication": false
  },
  "publication": {
    "authorized": false,
    "submitted": false
  }
}