LIVE_CLAIM_BINDING.json

{
  "bound_at": "2026-07-30T20:30:00+00:00",
  "claims": [
    "The optimal collaborative prediction set C*(x) has a two-threshold form: it augments the human-proposed set H(x) with labels whose score exceeds threshold a*, and prunes labels within H(x) whose score exceeds threshold b* (Theorem 2.1, Section 2).",
    "The counterfactual harm principle requires P(Y not in C(X) | Y in H(X)) < epsilon, i.e., the AI must not remove correct labels the human already proposed (Section 1).",
    "The complementarity principle requires P(Y in C(X) | Y not in H(X)) >= 1-delta, i.e., the AI must recover correct labels the human missed (Section 1).",
    "The offline calibration procedure gives distribution-free finite-sample coverage guarantees whose bounds depend on the sizes of the calibration sets used to fit a* and b* (Proposition 4.1, Section 4.1).",
    "The CUP-Online algorithm updates thresholds a_t and b_t via stochastic approximation steps (a_{t+1}=a_t+eta(1{s(x_t,y_t)>a_t}-delta), and analogously for b_t) to track target error rates under arbitrary distribution shift, including shifts caused by humans adapting their behavior to the AI (Section 4.2, Proposition 4.2)."
  ],
  "claims_dataset_sha256": "7c0373fbbfb98b5acdc2c0ac122d9d81431bd6b42dda76fa36a91b49ec4b7825",
  "claims_dataset_url": "https://icml-2026-agent-repro-challenge.static.hf.space/claims_anchored.json",
  "exact": true,
  "paper_id": "FzP6XZGG4d"
}