Pith Number

pith:HXOC35XR

pith:2026:HXOC35XRXCTLEKBO5GUMOC2X6Z

not attested not anchored not stored refs resolved

Learning to Decide with AI Assistance under Human-Alignment

Eleni Straitouri, Manuel Gomez-Rodriguez, Nina Corvelo Benz

Under perfect AI-human confidence alignment, expected regret for learning binary decisions drops to O(√(|H| T log T)).

arxiv:2605.12646 v1 · 2026-05-12 · cs.LG · cs.AI · cs.HC

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{HXOC35XRXCTLEKBO5GUMOC2X6Z}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Under perfect alignment between AI and human confidence, a learner can attain an expected regret of O(√(|H| · T log T)) and, when √|H| = O(log T) and B is countable, a non-trivial generalization of the Dvoretzky-Kiefer-Wolfowitz inequality improves the regret bound to O(√(T log T)).

C2weakest assumption

The assumption of perfect alignment between the sets of AI confidence values and human confidence values in their own predictions, which is required to reduce the effective state space and obtain the improved regret bounds.

C3one line summary

Alignment between AI and human confidence reduces the complexity of learning optimal binary decisions with AI assistance, lowering regret from Ω(√(|H|·|B|·T)) to O(√(|H|·T log T)) under perfect alignment.

References

18 extracted · 18 resolved · 0 Pith anchors

[1] Understanding the effect of accuracy on trust in machine learning models 2019

[2] Effect of confidence and explanation on accuracy and trust calibration in ai-assisted decision making 2020

[3] Towards optimiz- ing human-centric objectives in ai-assisted decision-making with offline reinforcement learning.arXiv preprint arXiv:2403.05911,

[4] Do humans trust advice more if it comes from ai? an analysis of human-ai interactions 2022

[5] From calibration to collaboration: Llm uncertainty quantification should be more human-centered.arXiv preprint arXiv:2506.07461,

Receipt and verification

First computed	2026-05-18T03:09:59.843992Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

3ddc2df6f1b8a6b2282ee9a8c70b57f670bf97c16952c90238218785eb31a274

Aliases

arxiv: 2605.12646 · arxiv_version: 2605.12646v1 · doi: 10.48550/arxiv.2605.12646 · pith_short_12: HXOC35XRXCTL · pith_short_16: HXOC35XRXCTLEKBO · pith_short_8: HXOC35XR

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/HXOC35XRXCTLEKBO5GUMOC2X6Z \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 3ddc2df6f1b8a6b2282ee9a8c70b57f670bf97c16952c90238218785eb31a274

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "97dd09543f176736a8ff9824eba3788454af9a16426117bce955f1b08d9c230d",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.HC"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-12T18:42:49Z",
    "title_canon_sha256": "10e3a5faeec4bde693f8866d95fdb9d0d45e14e79b21924c28107ff95781237a"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12646",
    "kind": "arxiv",
    "version": 1
  }
}