Pith Number

pith:V5LIG6OZ

pith:2026:V5LIG6OZF5PER336NLLXVNRM3S

not attested not anchored not stored refs resolved

PREPING: Building Agent Memory without Tasks

Jinheon Baek, Minki Kang, Sangwoo Park, Sung Ju Hwang, Yumin Choi

Agents can construct competitive procedural memory for new environments using only self-generated synthetic tasks before any real experience.

arxiv:2605.13880 v1 · 2026-05-11 · cs.AI · 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{V5LIG6OZF5PER336NLLXVNRM3S}

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

Preping substantially improves over a no-memory baseline and achieves performance competitive with strong playbook-based methods built from offline or online experience, with deployment cost $2.99× lower on AppWorld and $2.23× lower on BFCL v3 than online memory construction.

C2weakest assumption

That synthetic tasks generated and filtered by the proposer-validator loop will transfer to real target-environment tasks without any direct experience of those tasks.

C3one line summary

Preping builds agent memory via proposer-guided synthetic practice and selective validation, matching offline/online methods at 2-3x lower deployment cost.

References

53 extracted · 53 resolved · 7 Pith anchors

[1] Tool-r0: Self-evolving llm agents for tool-learning from zero data 2026 · doi:10.48550/arxiv.2602.21320

[2] Introducing the model context protocol, November 2024 2024

[3] DeepSeek-V3.2: Pushing the frontier of open large language models, 2025 2025

[4] LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation 2025 · doi:10.48550/arxiv.2510.04851

[5] R-Zero: Self-Evolving Reasoning LLM from Zero Data 2025 · doi:10.48550/arxiv.2508.05004

Receipt and verification

First computed	2026-05-17T23:39:19.197500Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714

Aliases

arxiv: 2605.13880 · arxiv_version: 2605.13880v1 · doi: 10.48550/arxiv.2605.13880 · pith_short_12: V5LIG6OZF5PE · pith_short_16: V5LIG6OZF5PER336 · pith_short_8: V5LIG6OZ

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/V5LIG6OZF5PER336NLLXVNRM3S \
  | 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: af568379d92f5e48ef7e6ad77ab62cdca2dbc4dc0147d965637b81954cb16714

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "cff4abde4570b51cd1946a9a209b315530b2fe54f925b569f88b961b7b457649",
    "cross_cats_sorted": [
      "cs.CL"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-11T04:34:43Z",
    "title_canon_sha256": "f33fac511cef95d8d740b902d062a69cd90aca34d371ab437bb74f51cd762dea"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13880",
    "kind": "arxiv",
    "version": 1
  }
}