pith. the verified trust layer for science. sign in
Pith Number

pith:63CLJGZZ

pith:2025:63CLJGZZRI6Y3TGVIEFSI6W2DD
not attested not anchored not stored refs resolved

Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks

Amir Gholami, Gopala Anumanchipalli, Hiroki Furuta, Kurt Keutzer, Lutfi Eren Erdogan, Nicholas Lee, Sehoon Kim, Suhong Moon

Plan-and-Act improves LLM agent performance on long-horizon tasks by separating planning from execution and training the planner with synthetic data.

arxiv:2503.09572 v3 · 2025-03-12 · cs.CL

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{63CLJGZZRI6Y3TGVIEFSI6W2DD}

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

We evaluate Plan-and-Act using web navigation as a representative long-horizon planning environment, demonstrating a state-of-the-art 57.58% success rate on the WebArena-Lite benchmark as well as a text-only state-of-the-art 81.36% success rate on WebVoyager.

C2weakest assumption

That annotating ground-truth trajectories with feasible plans and augmenting with diverse synthetic examples will produce plans that generalize to unseen tasks and environments without overfitting to the annotation process.

C3one line summary

Plan-and-Act trains a dedicated Planner on synthetic plan-annotated trajectories to generate high-level plans that an Executor follows, reaching 57.58% success on WebArena-Lite and 81.36% on WebVoyager.

References

101 extracted · 101 resolved · 15 Pith anchors

[1] Agent-e: From autonomous web navigation to foundational design principles in agentic systems.ArXiv, abs/2407.13032 2024
[2] Digirl: Training in-the-wild device-control agents with autonomous reinforcement learning 2024
[3] Web agents with world models: Learning and leveraging environment dynamics in web navigation 2024
[4] Mind2web: Towards a generalist agent for the web 2024
[5] E., Lee, N., Jha, S., Kim, S., Tabrizi, R., Moon, S., Hooper, C., Anumanchipalli, G., Keutzer, K., and Gholami, A 2024

Formal links

2 machine-checked theorem links

Cited by

17 papers in Pith

Receipt and verification
First computed 2026-05-17T23:38:13.022511Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

f6c4b49b398a3d8dccd5410b247ada18cd11833be778faa25494b09179006849

Aliases

arxiv: 2503.09572 · arxiv_version: 2503.09572v3 · doi: 10.48550/arxiv.2503.09572 · pith_short_12: 63CLJGZZRI6Y · pith_short_16: 63CLJGZZRI6Y3TGV · pith_short_8: 63CLJGZZ
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/63CLJGZZRI6Y3TGVIEFSI6W2DD \
  | 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: f6c4b49b398a3d8dccd5410b247ada18cd11833be778faa25494b09179006849
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "22c921f9cc8001db27ca5315ea28b031eece5810756a71c39c3e02c79adf1d3b",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-03-12T17:40:52Z",
    "title_canon_sha256": "e77e753965af9169f84a8430407792b6b1e43be96c8734ee8fc1b07e007e1c78"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2503.09572",
    "kind": "arxiv",
    "version": 3
  }
}