Pith Number

pith:X5R3JTR4

pith:2026:X5R3JTR4JBUIKUINLZUQCLYAVK

not attested not anchored not stored refs pending

Towards Efficient and Expressive Offline RL via Flow-Anchored Noise-conditioned Q-Learning

Dohyeong Kim, Eshan Balachandar, Keshav Pingali, Sungyoung Lee, Zelal Su Mustafaoglu

FAN achieves state-of-the-art offline RL performance using only a single flow-policy iteration and one Gaussian noise sample.

arxiv:2605.01663 v2 · 2026-05-03 · cs.LG · cs.RO

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{X5R3JTR4JBUIKUINLZUQCLYAVK}

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

Our theoretical analysis of convergence and performance bounds demonstrates that these simplifications not only improve efficiency but also lead to superior task performance. Experiments on robotic manipulation and locomotion tasks demonstrate that FAN achieves state-of-the-art performance while significantly reducing both training and inference runtimes.

C2weakest assumption

That a single flow-policy iteration plus one Gaussian noise sample for the distributional critic is sufficient to preserve both the expressivity of full iterative flows and the accuracy of multi-sample critics without introducing bias that the behavior-regularization term cannot correct.

C3one line summary

FAN achieves state-of-the-art offline RL performance on robotic tasks by anchoring flow policies and using single-sample noise-conditioned Q-learning, with proven convergence and reduced runtimes.

Receipt and verification

First computed	2026-05-29T01:05:11.368025Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

bf63b4ce3c486885510d5e69012f00aa8384bbf88b7c85a6cb17abb05c8d4ac1

Aliases

arxiv: 2605.01663 · arxiv_version: 2605.01663v2 · doi: 10.48550/arxiv.2605.01663 · pith_short_12: X5R3JTR4JBUI · pith_short_16: X5R3JTR4JBUIKUIN · pith_short_8: X5R3JTR4

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/X5R3JTR4JBUIKUINLZUQCLYAVK \
  | 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: bf63b4ce3c486885510d5e69012f00aa8384bbf88b7c85a6cb17abb05c8d4ac1

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "d45041ee201ca8c1a6bbe7d5376547b3aa0d0d90cd496f8364d355579af4f9ab",
    "cross_cats_sorted": [
      "cs.RO"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-03T01:32:11Z",
    "title_canon_sha256": "9163953be47922487f85a8383db863696a4148abbc3f1ffda5c63994e733a055"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.01663",
    "kind": "arxiv",
    "version": 2
  }
}