Pith Number

pith:24ZH7CZO

pith:2026:24ZH7CZOHOCKQI5UBKEGORMGTF

not attested not anchored not stored refs resolved

Multi-Objective and Mixed-Reward Reinforcement Learning via Reward-Decorrelated Policy Optimization

Jiahong Zhou, Jingang Wang, Kaiyuan Liu, Rongxiang Weng, Xin Chen, Xunliang Cai, Yang Bai, Ziyuan Zhuang

RDPO stabilizes advantages in mixed-reward reinforcement learning by normalizing magnitudes and removing correlations before aggregation.

arxiv:2605.13641 v1 · 2026-05-13 · cs.LG · cs.CL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{24ZH7CZOHOCKQI5UBKEGORMGTF}

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

When applied during the post-training of LongCat-Flash, RDPO enhances instruction following, writing quality, and robustness to hard prompts while remaining broadly competitive on reasoning and coding evaluations.

C2weakest assumption

That magnitude-aware quantile normalization and Mahalanobis whitening will stabilize advantages across heterogeneous rewards without discarding critical signal or introducing new biases in the specific reward distributions of the LongCat-Flash training setup.

C3one line summary

RDPO applies magnitude-aware quantile normalization and Mahalanobis whitening to decorrelate heterogeneous rewards in multi-objective RL, improving instruction following and writing quality on LongCat-Flash post-training while staying competitive on reasoning and coding.

References

17 extracted · 17 resolved · 9 Pith anchors

[1] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models · arXiv:2402.03300

[2] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models · doi:10.48550/arxiv.2402.03300

[3] GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization · doi:10.48550/arxiv.2601.05242

[4] Kimi k1.5: Scaling Reinforcement Learning with LLMs · doi:10.48550/arxiv.2501.12599

[5] L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning · doi:10.48550/arxiv.2503.04697

Receipt and verification

First computed	2026-05-18T02:44:17.603486Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

d7327f8b2e3b84a823b40a88674586996f0819a993e2e14038a9405da7ba7edf

Aliases

arxiv: 2605.13641 · arxiv_version: 2605.13641v1 · doi: 10.48550/arxiv.2605.13641 · pith_short_12: 24ZH7CZOHOCK · pith_short_16: 24ZH7CZOHOCKQI5U · pith_short_8: 24ZH7CZO

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/24ZH7CZOHOCKQI5UBKEGORMGTF \
  | 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: d7327f8b2e3b84a823b40a88674586996f0819a993e2e14038a9405da7ba7edf

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "89ad6306ebae2e291fe2e899775f6b3612ffad186d089f93b57cdd1eaf0ce81c",
    "cross_cats_sorted": [
      "cs.CL"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-13T15:05:18Z",
    "title_canon_sha256": "25c79d2d6d192456412abfcc88c8bcbce2ae575bf3285c8a57a47d72c4c76ca3"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13641",
    "kind": "arxiv",
    "version": 1
  }
}