pith. sign in
Pith Number

pith:QC25BQBJ

pith:2025:QC25BQBJB2YJ7IODOML6YMEXOC
not attested not anchored not stored refs resolved

MemEvolve: Meta-Evolution of Agent Memory Systems

Chong Zhan, Guibin Zhang, Haotian Ren, He Zhu, Junhao Wang, Shuicheng Yan, Wangchunshu Zhou, Zhenhong Zhou

MemEvolve lets LLM agents evolve both their stored experience and the memory architecture that organizes it.

arxiv:2512.18746 v1 · 2025-12-21 · cs.CL · cs.MA

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{QC25BQBJB2YJ7IODOML6YMEXOC}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more

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

MemEvolve achieves substantial performance gains, improving frameworks such as SmolAgent and Flash-Searcher by up to 17.06%, and demonstrates strong cross-task and cross-LLM generalization by designing transferable memory architectures.

C2weakest assumption

The modular decomposition of memory systems into encode/store/retrieve/manage components in EvolveLab is assumed to be expressive enough to support meaningful meta-evolution without omitting critical architectural variations present in real deployments.

C3one line summary

MemEvolve jointly evolves agent experiential knowledge and memory architectures via a modular codebase, delivering up to 17% gains on agent benchmarks with cross-task and cross-model generalization.

References

39 extracted · 39 resolved · 6 Pith anchors

[1] Pang, X., Peng, J., Peng, R., Qiao, Y., Qiu, J., Qu, X., Qu, Y., Ren, Y., Shang, F., Shao, W., Shen, J., Shen, S., Song, C., Song, D., Song, D., Su, C., Su, W., Sun, W., Sun, Y., Tan, Q., Tang, C., Ta 2025
[2] Cai, Y., Cai, S., Shi, Y., Xu, Z., Chen, L., Qin, Y., Tan, X., Li, G., Li, Z., Lin, H., Mao, Y., Li, K., and Sun, X. (2025). Training-free group relative policy optimization 2025
[3] Yin, Z., Ma, Z., and Mo, Z. (2025). xbench: Tracking agents productivity scaling with profession-aligned real-world evaluations 2025
[4] L., Ni, J., Xu, L., Li, M., Tian, N., Chen, R 2025
[5] Du, M., Xu, B., Zhu, C., Wang, X., and Mao, Z. (2025). Deepresearch bench: A comprehensive benchmark for deep research agents 2025

Formal links

3 machine-checked theorem links

Cited by

21 papers in Pith

Receipt and verification
First computed 2026-05-17T23:38:49.966891Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

80b5d0c0290eb09fa1c37317ec309770a51af5cbe95fd4b633139aaac439093c

Aliases

arxiv: 2512.18746 · arxiv_version: 2512.18746v1 · doi: 10.48550/arxiv.2512.18746 · pith_short_12: QC25BQBJB2YJ · pith_short_16: QC25BQBJB2YJ7IOD · pith_short_8: QC25BQBJ
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/QC25BQBJB2YJ7IODOML6YMEXOC \
  | 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: 80b5d0c0290eb09fa1c37317ec309770a51af5cbe95fd4b633139aaac439093c
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "76f53ab048b8ff64c74e2e4d88df62e9368db69dde90fabff38bbae5b769bd35",
    "cross_cats_sorted": [
      "cs.MA"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-12-21T14:26:14Z",
    "title_canon_sha256": "bd243b8466d2fabf124502b6b763299384c1e6bdd81ae8f7671f297a6e610adc"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2512.18746",
    "kind": "arxiv",
    "version": 1
  }
}