Pith Number

pith:OAXAX5EA

pith:2026:OAXAX5EABORLIHW6TSSGKGWTDU

not attested not anchored not stored refs resolved

SkillFlow: Flow-Driven Recursive Skill Evolution for Agentic Orchestration

Erik Cambria, Haoran Luo, Mingda Zhang, Tiesunlong Shen, Wenjin Liu, Xiaoying Tang, Zikai Xiao

SkillFlow uses tempered trajectory balance to sample reward-proportional strategies and drive autonomous recursive skill evolution in agent orchestration.

arxiv:2605.14089 v1 · 2026-05-13 · cs.AI

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{OAXAX5EABORLIHW6TSSGKGWTDU}

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

SkillFlow significantly outperforms baselines across question answering, mathematical reasoning, code generation, and real-world interactive decision making tasks on 14 datasets.

C2weakest assumption

That sampling trajectories proportional to reward via the tempered trajectory balance loss will reliably produce both diverse strategies and a backward policy that gives transparent per-step credit assignment without additional inference cost, and that the resulting flow diagnostics can autonomously drive correct skill creation and pruning decisions.

C3one line summary

SkillFlow trains agent orchestration policies with a regression-based flow-matching loss that preserves diverse strategies and drives autonomous skill creation and pruning.

References

13 extracted · 13 resolved · 0 Pith anchors

[1] For each s∈ S (k), compute the per-skill CGF Λ(s) λ at λ∈ {0,1} over the recent batch Bs via the zero-cost formulas of Proposition 20

[2] Derive the summaries G(s), Λ(s) 1 , and eΛ(s) = Λ (s) 1 −E s′[Λ(s′) 1 ] via Lemmas 22, 23 and Remark 7

[3] Classify eachs∈ S (k) intoD − k ,R k, orU k via Definition 14

[4] Refine eachs∈ U k viaΨin refine mode to produceU ′ k

[5] From the validation buffer, sample same-query success/failure pairs(τ +, τ −); identify trigger stepsT trig q via Definition 15

Receipt and verification

First computed	2026-05-17T23:39:12.234131Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

702e0bf4800ba2b41ede9ca4651ad31d09dde82dd9d40dd9609ae2be47fc2165

Aliases

arxiv: 2605.14089 · arxiv_version: 2605.14089v1 · doi: 10.48550/arxiv.2605.14089 · pith_short_12: OAXAX5EABORL · pith_short_16: OAXAX5EABORLIHW6 · pith_short_8: OAXAX5EA

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/OAXAX5EABORLIHW6TSSGKGWTDU \
  | 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: 702e0bf4800ba2b41ede9ca4651ad31d09dde82dd9d40dd9609ae2be47fc2165

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "30d51ed1808df9e8e2aa74f9b37195ded33f9a90658dfb4e669d695e62b498f0",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-13T20:14:44Z",
    "title_canon_sha256": "86aaf523f4452edb3693ace4aa1be71d372eb9b0eb242df1d96cb9ec3c5eb480"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14089",
    "kind": "arxiv",
    "version": 1
  }
}