Configuration: EvoFlows, per-family retrained Protocol: Ceiling-normalised comparison, EvoFlows published values, six families: Covariance (% of ceiling) Dataset: The six evaluation datasets behind EvoFlows' published figures 98.4% Covariance agreement (percent of the natural homolog ceiling)percent · higher
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows (Published), Covariance (% of ceiling), ceiling-normalised
Metrics normalised against the attainable ceiling computed on held-out natural homologs.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 4, row "Covariance (% of ceiling)", column "EvoFlows (Published)" Configuration: EvoFlows, per-family retrained Protocol: Ceiling-normalised comparison, EvoFlows published values, six families: Positional JS† (× floor) Dataset: The six evaluation datasets behind EvoFlows' published figures 1.8 Jensen shannon divergence (positional formulation chosen by the EditJumps authors, expressed as a multiple of the floor)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows (Published), Positional JS† (× floor), ceiling-normalised
Metrics normalised against the attainable ceiling computed on held-out natural homologs.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 4, row "Positional JS† (× floor)", column "EvoFlows (Published)" Configuration: EvoFlows, per-family retrained Protocol: Ceiling-normalised comparison, EvoFlows published values, six families: MIP (% of ceiling) Dataset: The six evaluation datasets behind EvoFlows' published figures 95.5% Mutual information product agreement (percent of the natural homolog ceiling)percent · higher
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows (Published), MIP (% of ceiling), ceiling-normalised
Metrics normalised against the attainable ceiling computed on held-out natural homologs.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 4, row "MIP (% of ceiling)", column "EvoFlows (Published)" Configuration: EvoFlows, per-family retrained Protocol: Ceiling-normalised comparison, EvoFlows published values, six families: Pairwise Distance‡ Dataset: The six evaluation datasets behind EvoFlows' published figures 1 Levenshtein distance (pooled reduction, expressed as a ratio to natural homologs)unitless · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows (Published), Pairwise Distance‡, ceiling-normalised
Metrics normalised against the attainable ceiling computed on held-out natural homologs.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 4, row "Pairwise Distance‡", column "EvoFlows (Published)" Configuration: EvoDiff-MSA, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 13.8 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoDiff-MSA on HER2-VH, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoDiff-MSA", column "edits/seq", sub-column "HER2" Configuration: EvoDiff-MSA, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 4.43 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoDiff-MSA on Ty1, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoDiff-MSA", column "edits/seq", sub-column "Ty1" Configuration: EvoFlows, per-family retrained Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 35.3 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows on HER2-VH, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoFlows", column "edits/seq", sub-column "HER2" Configuration: EvoFlows, per-family retrained Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 14.6 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows on Ty1, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoFlows", column "edits/seq", sub-column "Ty1" Configuration: Forced-substitution evotuned language model, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 38.7 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning (forced) on HER2-VH, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning (forced)", column "edits/seq", sub-column "HER2" Configuration: Forced-substitution evotuned language model, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 15.3 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning (forced) on Ty1, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning (forced)", column "edits/seq", sub-column "Ty1" Configuration: Evotuned language model, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 11.8 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning on HER2-VH, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning", column "edits/seq", sub-column "HER2" Configuration: Evotuned language model, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 2.54 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning on Ty1, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning", column "edits/seq", sub-column "Ty1" Configuration: Random mutation floor, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 35.1 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source random mutations on HER2-VH, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "random mutations", column "edits/seq", sub-column "HER2" Configuration: Random mutation floor, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 13.9 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source random mutations on Ty1, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "random mutations", column "edits/seq", sub-column "Ty1" Configuration: Random homolog pairing ceiling, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 93.9 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source random pairing on HER2-VH, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "random pairing", column "edits/seq", sub-column "HER2" Configuration: Random homolog pairing ceiling, published values Protocol: EvoFlows published values, two families: edits/seq Dataset: The six evaluation datasets behind EvoFlows' published figures 30.5 Edit count (mean edits per sequence as the EvoFlows paper prints it, under conditions the replication could not recover)count · unknown
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source random pairing on Ty1, edits/seq, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "random pairing", column "edits/seq", sub-column "Ty1" Configuration: EvoDiff-MSA, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 10.2 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoDiff-MSA on HER2-VH, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoDiff-MSA", column "KL ×103", sub-column "HER2" Configuration: EvoDiff-MSA, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 11.6 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoDiff-MSA on Ty1, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoDiff-MSA", column "KL ×103", sub-column "Ty1" Configuration: EvoFlows, per-family retrained Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 23.2 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows on HER2-VH, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoFlows", column "KL ×103", sub-column "HER2" Configuration: EvoFlows, per-family retrained Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 21.5 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source EvoFlows on Ty1, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "EvoFlows", column "KL ×103", sub-column "Ty1" Configuration: Forced-substitution evotuned language model, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 36.5 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning (forced) on HER2-VH, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning (forced)", column "KL ×103", sub-column "HER2" Configuration: Forced-substitution evotuned language model, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 54.8 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning (forced) on Ty1, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning (forced)", column "KL ×103", sub-column "Ty1" Configuration: Evotuned language model, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 10.7 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning on HER2-VH, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning", column "KL ×103", sub-column "HER2" Configuration: Evotuned language model, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 10.4 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source evotuning on Ty1, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "evotuning", column "KL ×103", sub-column "Ty1" Configuration: Random mutation floor, published values Protocol: EvoFlows published values, two families: KL x10^3 Dataset: The six evaluation datasets behind EvoFlows' published figures 86.7 Kullback leibler divergence (pooled amino-acid composition, under conditions the replication could not recover)unitless · lower
Uncertainty: Not reported by the source: The table prints a single value with no interval.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The paper states in section 4 that its computed values and the published EvoFlows values are not comparable and that it could not recover the original experimental conditions from the EvoFlows paper, and Table 1's caption states that the two blocks are not comparable. The records carrying EvoFlows' own published values are therefore withheld from automatic comparison with the reimplemented block, and neither set may be pooled into one leaderboard.When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Section 4, first sentence; Table 1 caption; Table 4 caption; Table 5
Methods, coverage and source random mutations on HER2-VH, KL x10^3, EvoFlows published values
The conditions under which the EvoFlows paper produced these values, which the replication states it could not recover.
Aggregation: Not reported
When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows (arXiv v1) · Table 1, lower block, row "random mutations", column "KL ×103", sub-column "HER2"