rewirebio.iobenchmarks
Source

PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256

Original source

What this source contributed

Records in this release that cite this source. A record can cite several sources.

Results

Values as the source printed them. Compare results only under compatible protocols, datasets and metrics.

Model or configurationMetricValueDatasetEvaluation
Biolord*Perturbench cosine logfc rank0.027 ± 1 × 10 − 3 fractionNorman19 (PerturBench split)Biolord* on PerturBench CB-COSINE-RANK: combination prediction on Norman19, Cosine LogFC rank
Biolord*Perturbench cosine logfc0.41 ± 2 × 10 − 2 fractionNorman19 (PerturBench split)Biolord* on PerturBench CB-COSINE: combination prediction on Norman19, Cosine similarity of log fold change
Biolord*Perturbench RMSE rank0.028 ± 1 × 10 − 3 fractionNorman19 (PerturBench split)Biolord* on PerturBench CB-RMSE-RANK: combination prediction on Norman19, RMSE mean rank
Biolord*Perturbench RMSE of mean0.086 ± 6 × 10 − 4 unit-unreportedNorman19 (PerturBench split)Biolord* on PerturBench CB-RMSE: combination prediction on Norman19, RMSE of the mean
Biolord*Perturbench cosine logfc rank0.37 ± 2 × 10 − 2 fractionSrivatsan20 (PerturBench split)Biolord* on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
Biolord*Perturbench cosine logfc0.18 ± 1 × 10 − 1 fractionSrivatsan20 (PerturBench split)Biolord* on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change
Biolord*Perturbench RMSE rank0.35 ± 1 × 10 − 1 fractionSrivatsan20 (PerturBench split)Biolord* on PerturBench CT-RMSE-RANK: covariate transfer on Srivatsan20, RMSE mean rank
Biolord*Perturbench RMSE of mean0.086 ± 4 × 10 − 2 unit-unreportedSrivatsan20 (PerturBench split)Biolord* on PerturBench CT-RMSE: covariate transfer on Srivatsan20, RMSE of the mean
CPA* (noAdv)Perturbench cosine logfc rank0.16 ± 4 × 10 − 2 fractionNorman19 (PerturBench split)CPA* (noAdv) on PerturBench CB-COSINE-RANK: combination prediction on Norman19, Cosine LogFC rank
CPA* (noAdv)Perturbench cosine logfc0.55 ± 9 × 10 − 2 fractionNorman19 (PerturBench split)CPA* (noAdv) on PerturBench CB-COSINE: combination prediction on Norman19, Cosine similarity of log fold change
CPA* (noAdv)Perturbench RMSE rank0.18 ± 2 × 10 − 2 fractionNorman19 (PerturBench split)CPA* (noAdv) on PerturBench CB-RMSE-RANK: combination prediction on Norman19, RMSE mean rank
CPA* (noAdv)Perturbench RMSE of mean0.073 ± 6 × 10 − 3 unit-unreportedNorman19 (PerturBench split)CPA* (noAdv) on PerturBench CB-RMSE: combination prediction on Norman19, RMSE of the mean
CPA* (noAdv)Perturbench cosine logfc rank0.33 ± 3 × 10 − 2 fractionSrivatsan20 (PerturBench split)CPA* (noAdv) on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
CPA* (noAdv)Perturbench cosine logfc0.37 ± 4 × 10 − 2 fractionSrivatsan20 (PerturBench split)CPA* (noAdv) on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change
CPA* (noAdv)Perturbench RMSE rank0.29 ± 7 × 10 − 3 fractionSrivatsan20 (PerturBench split)CPA* (noAdv) on PerturBench CT-RMSE-RANK: covariate transfer on Srivatsan20, RMSE mean rank
CPA* (noAdv)Perturbench RMSE of mean0.020 ± 8 × 10 − 4 unit-unreportedSrivatsan20 (PerturBench split)CPA* (noAdv) on PerturBench CT-RMSE: covariate transfer on Srivatsan20, RMSE of the mean
CPA* (scGPT)Perturbench cosine logfc rank0.064 ± 1 × 10 − 2 fractionNorman19 (PerturBench split)CPA* (scGPT) on PerturBench CB-COSINE-RANK: combination prediction on Norman19, Cosine LogFC rank
CPA* (scGPT)Perturbench cosine logfc0.70 ± 1 × 10 − 2 fractionNorman19 (PerturBench split)CPA* (scGPT) on PerturBench CB-COSINE: combination prediction on Norman19, Cosine similarity of log fold change
CPA* (scGPT)Perturbench RMSE rank0.13 ± 2 × 10 − 2 fractionNorman19 (PerturBench split)CPA* (scGPT) on PerturBench CB-RMSE-RANK: combination prediction on Norman19, RMSE mean rank
CPA* (scGPT)Perturbench RMSE of mean0.061 ± 2 × 10 − 3 unit-unreportedNorman19 (PerturBench split)CPA* (scGPT) on PerturBench CB-RMSE: combination prediction on Norman19, RMSE of the mean
CPA* (scGPT)Perturbench cosine logfc rank0.38 ± 2 × 10 − 2 fractionSrivatsan20 (PerturBench split)CPA* (scGPT) on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
CPA* (scGPT)Perturbench cosine logfc0.29 ± 9 × 10 − 4 fractionSrivatsan20 (PerturBench split)CPA* (scGPT) on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change
CPA* (scGPT)Perturbench RMSE rank0.32 ± 1 × 10 − 2 fractionSrivatsan20 (PerturBench split)CPA* (scGPT) on PerturBench CT-RMSE-RANK: covariate transfer on Srivatsan20, RMSE mean rank
CPA* (scGPT)Perturbench RMSE of mean0.021 ± 3 × 10 − 4 unit-unreportedSrivatsan20 (PerturBench split)CPA* (scGPT) on PerturBench CT-RMSE: covariate transfer on Srivatsan20, RMSE of the mean
CPA*Perturbench cosine logfc rank0.12 ± 2 × 10 − 2 fractionNorman19 (PerturBench split)CPA* on PerturBench CB-COSINE-RANK: combination prediction on Norman19, Cosine LogFC rank
CPA*Perturbench cosine logfc0.52 ± 6 × 10 − 2 fractionNorman19 (PerturBench split)CPA* on PerturBench CB-COSINE: combination prediction on Norman19, Cosine similarity of log fold change
CPA*Perturbench RMSE rank0.17 ± 3 × 10 − 2 fractionNorman19 (PerturBench split)CPA* on PerturBench CB-RMSE-RANK: combination prediction on Norman19, RMSE mean rank
CPA*Perturbench RMSE of mean0.079 ± 5 × 10 − 3 unit-unreportedNorman19 (PerturBench split)CPA* on PerturBench CB-RMSE: combination prediction on Norman19, RMSE of the mean
CPA*Perturbench cosine logfc rank0.35 ± 6 × 10 − 3 fractionSrivatsan20 (PerturBench split)CPA* on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
CPA*Perturbench cosine logfc0.31 ± 1 × 10 − 2 fractionSrivatsan20 (PerturBench split)CPA* on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change
CPA*Perturbench RMSE rank0.32 ± 7 × 10 − 3 fractionSrivatsan20 (PerturBench split)CPA* on PerturBench CT-RMSE-RANK: covariate transfer on Srivatsan20, RMSE mean rank
CPA*Perturbench RMSE of mean0.021 ± 4 × 10 − 4 unit-unreportedSrivatsan20 (PerturBench split)CPA* on PerturBench CT-RMSE: covariate transfer on Srivatsan20, RMSE of the mean
Decoder (Cov)Perturbench cosine logfc rank0.47 ± 9 × 10 − 3 fractionSrivatsan20 (PerturBench split)Decoder (Cov) on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
Decoder (Cov)Perturbench cosine logfc0.30 ± 1 × 10 − 2 fractionSrivatsan20 (PerturBench split)Decoder (Cov) on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change
Decoder (Cov)Perturbench RMSE rank0.50 ± 4 × 10 − 2 fractionSrivatsan20 (PerturBench split)Decoder (Cov) on PerturBench CT-RMSE-RANK: covariate transfer on Srivatsan20, RMSE mean rank
Decoder (Cov)Perturbench RMSE of mean0.023 ± 3 × 10 − 5 unit-unreportedSrivatsan20 (PerturBench split)Decoder (Cov) on PerturBench CT-RMSE: covariate transfer on Srivatsan20, RMSE of the mean
DecoderPerturbench cosine logfc rank0.017 ± 6 × 10 − 3 fractionNorman19 (PerturBench split)Decoder on PerturBench CB-COSINE-RANK: combination prediction on Norman19, Cosine LogFC rank
DecoderPerturbench cosine logfc0.73 ± 2 × 10 − 2 fractionNorman19 (PerturBench split)Decoder on PerturBench CB-COSINE: combination prediction on Norman19, Cosine similarity of log fold change
DecoderPerturbench RMSE rank0.014 ± 4 × 10 − 4 fractionNorman19 (PerturBench split)Decoder on PerturBench CB-RMSE-RANK: combination prediction on Norman19, RMSE mean rank
DecoderPerturbench RMSE of mean0.043 ± 3 × 10 − 4 unit-unreportedNorman19 (PerturBench split)Decoder on PerturBench CB-RMSE: combination prediction on Norman19, RMSE of the mean
DecoderPerturbench cosine logfc rank0.16 ± 1 × 10 − 2 fractionSrivatsan20 (PerturBench split)Decoder on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
DecoderPerturbench cosine logfc0.35 ± 5 × 10 − 3 fractionSrivatsan20 (PerturBench split)Decoder on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change
DecoderPerturbench RMSE rank0.14 ± 7 × 10 − 3 fractionSrivatsan20 (PerturBench split)Decoder on PerturBench CT-RMSE-RANK: covariate transfer on Srivatsan20, RMSE mean rank
DecoderPerturbench RMSE of mean0.018 ± 1 × 10 − 4 unit-unreportedSrivatsan20 (PerturBench split)Decoder on PerturBench CT-RMSE: covariate transfer on Srivatsan20, RMSE of the mean
LA (scGPT)Perturbench cosine logfc rank0.0085 ± 1 × 10 − 3 fractionNorman19 (PerturBench split)LA (scGPT) on PerturBench CB-COSINE-RANK: combination prediction on Norman19, Cosine LogFC rank
LA (scGPT)Perturbench cosine logfc0.77 ± 4 × 10 − 3 fractionNorman19 (PerturBench split)LA (scGPT) on PerturBench CB-COSINE: combination prediction on Norman19, Cosine similarity of log fold change
LA (scGPT)Perturbench RMSE rank0.013 ± 2 × 10 − 3 fractionNorman19 (PerturBench split)LA (scGPT) on PerturBench CB-RMSE-RANK: combination prediction on Norman19, RMSE mean rank
LA (scGPT)Perturbench RMSE of mean0.044 ± 4 × 10 − 4 unit-unreportedNorman19 (PerturBench split)LA (scGPT) on PerturBench CB-RMSE: combination prediction on Norman19, RMSE of the mean
LA (scGPT)Perturbench cosine logfc rank0.13 ± 7 × 10 − 3 fractionSrivatsan20 (PerturBench split)LA (scGPT) on PerturBench CT-COSINE-RANK: covariate transfer on Srivatsan20, Cosine LogFC rank
LA (scGPT)Perturbench cosine logfc0.50 ± 4 × 10 − 3 fractionSrivatsan20 (PerturBench split)LA (scGPT) on PerturBench CT-COSINE: covariate transfer on Srivatsan20, Cosine similarity of log fold change

Claims

Statements this source supports, each with where in the source it was checked.

Models, benchmarks and other records

Configurations (10)

Benchmarks (1)

Tasks (8)

Dataset subsets (2)

Evaluations (76)

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

15 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-c8bea25e5abb
Property and statementOriginal source and locationReview and provenance
Artifact SHA-256
5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.artifact_sha256

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Artifact url
https://arxiv.org/html/2408.10609v1
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.artifact_url

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Hash scope
Exact retrieved primary paper artifact bytes.
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.hash_scope

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Retrieved at
2026-09-17T08:06:28.400469+00:00
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.retrieved_at

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Review method
1 values
  • automated-source-review
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.review.method

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Review: scope
Source identification and table transcription; no independent experimental reproduction.
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.review.scope

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Url
https://arxiv.org/html/2408.10609v1
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.url

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Version
Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: attributes.version

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Description
Primary reference; inspect retrieval status and review scope.
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: description

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

Search facets
None recorded
Source metadata
PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Original source ↗

No field-specific location recorded

Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Retrieved: 2026-09-17T08:06:28.400469+00:00

catalogued

No individual claim review recorded

Audit details

Field: facets

Source artifact SHA-256: 5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186

Hash scope: Exact retrieved primary paper artifact bytes.

Inspected artifact

How to read the evidence

Sources and history

Release 2026-11-21-c8bea25e5abb · Record review: source checked

Primary reference; inspect retrieval status and review scope.

Release history

No supporting source is linked yet.

Read original source

Download this release (gzip)
Technical metadata and extraction receipts

Stable ID: evidence-expansion-perturbench-5c480456

url
https://arxiv.org/html/2408.10609v1
version
Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
retrieved at
2026-09-17T08:06:28.400469+00:00
artifact sha256
5c4804565dd9faa17a11853a79e9847dcb6da73715b4c91f62e5b274cc79f186
artifact url
https://arxiv.org/html/2408.10609v1
hash scope
Exact retrieved primary paper artifact bytes.
review
scope: Source identification and table transcription; no independent experimental reproduction.; method: automated-source-review
Related records

    Suggest a correction