Pith Number

pith:Q6GKIYRW

pith:2026:Q6GKIYRWFRYSTWSVFE3UNRYKEX

not attested not anchored not stored refs resolved

Discrete Diffusion for Complex and Congested Multi-Agent Path Finding with Sparse Social Attention

Hongguang Wang, Jiaming Guo, Shiyu Quan, Tian Zhi, Xing Hu, Yang Zhao, Yuanzhe Wang, Yunji Chen, Zidong Du, Zihang Wei, Zisheng Liu

A discrete diffusion model generates initial joint plans that let a repair solver achieve 95.8 percent success on crowded multi-agent path problems with hundreds of agents.

arxiv:2605.13296 v1 · 2026-05-13 · cs.AI · cs.LG · cs.MA

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{Q6GKIYRWFRYSTWSVFE3UNRYKEX}

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

Across 20 complex and congested settings, DiffLNS achieves an average success rate of 95.8%, outperforming the strongest tested baseline by 9.6 percentage points and matching or exceeding all baselines in all 20 settings. The initializer generalizes to scenarios with up to 312 agents at inference time despite training only on instances with at most 96 agents.

C2weakest assumption

The spatiotemporal prior learned by the discrete diffusion model from expert demonstrations on small instances remains useful and produces repair-friendly drafts when applied to much larger agent counts and unseen congested layouts.

C3one line summary

DiffLNS uses a discrete diffusion initializer to produce warm-start plans that lift LNS2 success rates to 95.8% across 20 congested MAPF settings, generalizing from 96 to 312 agents.

References

26 extracted · 26 resolved · 1 Pith anchors

[1] URL https://ojs.aaai.org/index.php/ AAAI/article/view/34477 · doi:10.1609/aaai.v39i22.34477

[2] URL https://proceedings.neurips.cc/paper_files/paper/ 2021/file/958c530554f78bcd8e97125b70e6973d-Paper.pdf. J. Carvalho, A. T. Le, M. Baierl, D. Koert, and J. Peters. Motion planning diffusion: Learni 2021

[3] Le, Mark Baierl, Dorothea Koert, and Jan Peters 1986

[4] doi: 10.1109/ROBOT.1986. 1087401. Emiel Hoogeboom, Didrik Nielsen, Priyank Jaini, Patrick Forré, and Max Welling. Argmax flows and multino- mial diffusion: learning categorical distributions. InProcee 1986 · doi:10.1109/robot.1986

[5] URL https://ojs.aaai.org/index.php/ AAAI/article/view/21168 2024 · doi:10.1609/aaai.v36i9.21168

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-18T02:44:49.080261Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

878ca462362c7129da55293746c70a25dce9381ac8fe3d056d2d93c123378229

Aliases

arxiv: 2605.13296 · arxiv_version: 2605.13296v1 · doi: 10.48550/arxiv.2605.13296 · pith_short_12: Q6GKIYRWFRYS · pith_short_16: Q6GKIYRWFRYSTWSV · pith_short_8: Q6GKIYRW

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/Q6GKIYRWFRYSTWSVFE3UNRYKEX \
  | 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: 878ca462362c7129da55293746c70a25dce9381ac8fe3d056d2d93c123378229

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "887e81d9f074ec4f49a9515d145205e84fb52f2de807155aafbf9c97ffb311ad",
    "cross_cats_sorted": [
      "cs.LG",
      "cs.MA"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-13T10:10:22Z",
    "title_canon_sha256": "86da8987df118a32f7575fe174197dd60e3b9dcb1b2ae3ab7b8b848047278994"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13296",
    "kind": "arxiv",
    "version": 1
  }
}