pith:HKCEZK66
OGER: A Robust Offline-Guided Exploration Reward for Hybrid Reinforcement Learning
The OGER framework improves LLM reasoning by integrating offline guidance with an entropy-based exploration reward in hybrid reinforcement learning.
arxiv:2604.18530 v2 · 2026-04-20 · cs.AI
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{HKCEZK66GYSVJHWR6PV26EC3MB}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
OGER significantly outperforms competitive baselines, achieving substantial gains in mathematical reasoning while maintaining robust generalization to out-of-domain tasks.
That the entropy-aware reward modulation, when combined with multi-teacher offline guidance, reliably incentivizes useful exploration rather than noise or overfitting to the offline dataset.
OGER adds an auxiliary exploration reward built from offline trajectories and model entropy to hybrid RL training, yielding gains on math reasoning benchmarks and out-of-domain generalization.
Receipt and verification
| First computed | 2026-05-28T01:04:40.487248Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
3a844cabde3625549ed1f3ebaf105b6074329d4a593e2d39d8452f3ae09b7201
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/HKCEZK66GYSVJHWR6PV26EC3MB \
| 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: 3a844cabde3625549ed1f3ebaf105b6074329d4a593e2d39d8452f3ae09b7201
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "e4bf4d3f31f43abc77765d6a6fd84b6f8c25a7eb278d302b1b6d86f25776f9ad",
"cross_cats_sorted": [],
"license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
"primary_cat": "cs.AI",
"submitted_at": "2026-04-20T17:26:00Z",
"title_canon_sha256": "ff665f7e1cb0a558f475abc79644fdadc3f2483e62ea5a6746f0621f18c857e2"
},
"schema_version": "1.0",
"source": {
"id": "2604.18530",
"kind": "arxiv",
"version": 2
}
}