pith:FY7XKCPD
Curiosity-Critic: Cumulative Prediction Error Improvement as a Tractable Intrinsic Reward for World Model Training
Curiosity-Critic uses cumulative prediction error improvement as an intrinsic reward for world model training, estimated via a co-trained critic.
arxiv:2604.18701 v3 · 2026-04-20 · cs.LG · cs.AI · stat.ML
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{FY7XKCPDIB2RFDHSMQO75EI2PF}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Experiments on a stochastic grid world show that Curiosity-Critic outperforms prediction-error, visitation-count, and Random Network Distillation methods in training speed and final world model accuracy.
The learned critic converges well before the world model saturates, providing a reliable online estimate of the asymptotic error baseline without oracle knowledge of the noise floor.
Curiosity-Critic rewards the improvement in cumulative prediction error via a tractable per-step surrogate (current error minus learned asymptotic baseline), outperforming prior curiosity methods in a stochastic grid world.
Cited by
Receipt and verification
| First computed | 2026-06-19T16:09:58.264685Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
2e3f7509e34075128cf2641dfe911a7969e38e3d397a8d6b2af24b7a6bd55575
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/FY7XKCPDIB2RFDHSMQO75EI2PF \
| 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: 2e3f7509e34075128cf2641dfe911a7969e38e3d397a8d6b2af24b7a6bd55575
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "02cb1f8bece5d1793abd1b5ac478bf11832fb7f3098d349cc55835e5e8912221",
"cross_cats_sorted": [
"cs.AI",
"stat.ML"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.LG",
"submitted_at": "2026-04-20T18:01:15Z",
"title_canon_sha256": "6168d3d9a9a87b910e40adb8d3458536f23bc9972783af382d101ed5f2e0bb99"
},
"schema_version": "1.0",
"source": {
"id": "2604.18701",
"kind": "arxiv",
"version": 3
}
}