Configuration: AMPLIFY-350M (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.579 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source AMPLIFY-350M on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'AMPLIFY-350M' (page 12 of 13) Configuration: Ankh-base (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.649 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source Ankh-base on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 7, row 'Stability', column 'Ankh-base' (page 13 of 13) Configuration: Ankh3-large (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.759 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source Ankh3-large on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 7, row 'Stability', column 'Ankh3-large' (page 13 of 13) Configuration: DPLM-650M (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.746 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source DPLM-650M on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'DPLM-650M' (page 12 of 13) Configuration: E1-600M (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.712 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source E1-600M on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'E1-600M' (page 12 of 13) Configuration: ESM-1b (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.596 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source ESM-1b on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'ESM-1b' (page 12 of 13) Configuration: ESM-2 650M (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.639 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source ESM-2 650M on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'ESM-2 650M' (page 12 of 13) Configuration: ESM C 600M (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.706 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source ESM C 600M on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'ESM C 600M' (page 12 of 13) Configuration: ESM3-open (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.758 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source ESM3-open on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 7, row 'Stability', column 'ESM3-open' (page 13 of 13) Configuration: HyPLM-Flash, BF16 evaluation (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.687 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source HyPLM-Flash on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 6, row 'Stability', column 'HyPLM-Flash' (page 12 of 13) Configuration: ProtBERT (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.516 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source ProtBERT on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 7, row 'Stability', column 'ProtBERT' (page 13 of 13) Configuration: ProtT5-XL (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.586 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source ProtT5-XL on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 7, row 'Stability', column 'ProtT5-XL' (page 13 of 13) Configuration: xTrimoPGLM-1B (HyPLM report) Protocol: Stability (HyPLM report) Dataset: Stability (TAPE, HyPLM report) 0.66 Spearman (Spearman correlation on the TAPE stability regression split)unitless · higher
Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.
Coverage: Not reported scored / Not reported eligible
Author-reported evaluation · Source checked Evidence concern: excluded from comparisons The recent-PDB likelihood panel is described two ways. Section 4.3 and the Table 3 caption give 887 PDB chains released after 2 September 2026; Table 4 gives 887 RCSB PDB entries released 9 September to 7 October 2026. The start of the window and whether the unit is a chain or an entry both differ, so the panel cannot be rebuilt from the report.HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.3; Table 3 caption; Table 4 row 'PDB time split'
Appendix A states that the authors could not reproduce xTrimoPGLM's published ProteinGym scores (their runs fall 0.07 to 0.09 below the leaderboard) and that its likelihood is implausibly high, pointing to an incompatibility between their adapter and its masked-language-model head. They withhold its zero-shot and likelihood results and keep its embedding-based results as unaffected, but those results come from the same adapter and that claim was not tested.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Deviations'
Appendix A states that for GB1 two-vs-rest, GB1 three-vs-rest and the cross-protein hydrolase split of FLIP2, validation data come from the training distribution while the test sets are strongly shifted, so probe selection is dominated by noise and numerically equivalent runs of the same model differ by up to 0.2. The authors print these cells but leave them out of every summary statistic.HyPLM Technical Report (Hyra Team, 2026-10-12) · Appendix A 'Statistics'
Methods, coverage and source xTrimoPGLM-1B on stability (HyPLM report)
hyplm-flash-20261106-protocol-stability
Aggregation: Not reported
HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 7, row 'Stability', column 'xTrimoPGLM-1B' (page 13 of 13)