Pith Number

pith:OG7IS5F3

pith:2026:OG7IS5F32YRFHAAHZUPDK35DSA

not attested not anchored not stored refs resolved

Pretrained Model Representations as Acquisition Signals for Active Learning of MLIPs

Eszter Varga-Umbrich, Jules Tilly, Olivier Peltre, Paul Duckworth, Shikha Surana, Zachary Weller-Davies

Pretrained MLIP latent spaces supply acquisition signals that reduce active learning data needs for reactive chemistry.

arxiv:2605.03964 v2 · 2026-05-05 · cs.LG · physics.chem-ph

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{OG7IS5F32YRFHAAHZUPDK35DSA}

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

On reactive-chemistry benchmarks, both kernels consistently outperform fixed-descriptor baselines, committee disagreement, and random acquisition, reducing the data required to reach performance targets by an average of 38% for energy error and 28% for force error.

C2weakest assumption

The latent space of a pretrained MLIP already contains the information necessary for effective acquisition, eliminating the need for auxiliary uncertainty heads, Bayesian training and fine-tuning, or committee ensembles.

C3one line summary

Kernels from pretrained MLIP latent spaces outperform standard acquisition methods in active learning for reactive chemistry, reducing required labels by 38% for energy error and 28% for force error.

References

109 extracted · 109 resolved · 1 Pith anchors

[2] On representing chemical environments 2013 · doi:10.1103/physrevb.87.184115

[4] A foundation model for atomistic materials chemistry 2023 · arXiv:2401.00096

[5] doi:10.48550/arXiv.2206.07697 , title = 2023

[6] L., Marsalek, O., and Schran, C 2025 · doi:10.1063/5.0288994

[7] Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces 2007 · doi:10.1103/physrevlett.98.146401

Formal links

2 machine-checked theorem links

Cited by

2 papers in Pith

Force-Aware Neural Tangent Kernels for Scalable and Robust Active Learning of MLIPs

Receipt and verification

First computed	2026-05-20T00:01:42.689244Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

71be8974bbd622538007cd1e356fa39032960fec5cd844035b7efae252557132

Aliases

arxiv: 2605.03964 · arxiv_version: 2605.03964v2 · doi: 10.48550/arxiv.2605.03964 · pith_short_12: OG7IS5F32YRF · pith_short_16: OG7IS5F32YRFHAAH · pith_short_8: OG7IS5F3

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/OG7IS5F32YRFHAAHZUPDK35DSA \
  | 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: 71be8974bbd622538007cd1e356fa39032960fec5cd844035b7efae252557132

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "c503f41439c75f83afd823e0a93b25e131e0f850d3fb8e02b1351b595578f1f5",
    "cross_cats_sorted": [
      "physics.chem-ph"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-05T16:48:23Z",
    "title_canon_sha256": "59fad99b6b837a93829fb4fb40e66fe22db617b0c746ed269a7acde81b7499ce"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.03964",
    "kind": "arxiv",
    "version": 2
  }
}