# Pronoun Disambiguation on PDP-60 Source: [Pronoun Disambiguation on PDP-60]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:27:27.941Z Evaluation of model performance on the Pronoun Disambiguation task using the PDP-60 dataset, typically measured by Accuracy. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Pronoun Disambiguation | | Dataset (including any stated split) | PDP-60 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Maximum Attention Score (MAS) | 68.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | USSM + Supervised Deepnet + 3 Knowledge Bases | 66.7 | Not recorded | 68fa65dc1e209131cea0ce64 | | Single LM (full) | 60 | Not recorded | 68fa65dc1e209131cea0ce64 | | Subword-level Transformer LM | 58.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | Quan Liu | 58.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | USSM + Cause-Effect + WordNet + ConceptNet KB | 56.7 | Not recorded | 68fa65dc1e209131cea0ce64 | | USSM + Cause-Effect Knowledge Base | 55 | Not recorded | 68fa65dc1e209131cea0ce64 | | Single LM (partial) | 53.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | USSM + Supervised Deepnet | 53.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | Unsupervised Semantic Similarity Method (USSM) | 48.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | Nicos Issak | 48.3 | Not recorded | 68fa65dc1e209131cea0ce64 | | Patric Dhondt | 45 | Not recorded | 68fa65dc1e209131cea0ce64 | ### Maximum Attention Score (MAS) Accuracy: 68.3. Result ID: 698cce008b517eb4331cdf02. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"supervised","backbone":"BERT"}. This record does not assert independent reproduction. ### USSM + Supervised Deepnet + 3 Knowledge Bases Accuracy: 66.7. Result ID: 698cce008b517eb4331cdf01. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"supervised","knowledge\_base":"3 Knowledge Bases"}. This record does not assert independent reproduction. ### Single LM (full) Accuracy: 60. Result ID: 698cce008b517eb4331cdefc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"unsupervised","variant":"full"}. This record does not assert independent reproduction. ### Subword-level Transformer LM Accuracy: 58.3. Result ID: 698cce008b517eb4331cdefa. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"unsupervised"}. This record does not assert independent reproduction. ### Quan Liu Accuracy: 58.3. Result ID: 698cce008b517eb4331cdeff. [Reporting paper]() Recorded evaluation configuration: {"supervision":"supervised","competition":"WS Challenge 2016 - winner"}. This record does not assert independent reproduction. ### USSM + Cause-Effect + WordNet + ConceptNet KB Accuracy: 56.7. Result ID: 698cce008b517eb4331cdef9. [Reporting paper]() Recorded evaluation configuration: {"supervision":"unsupervised","knowledge\_base":"Cause-Effect, WordNet, ConceptNet"}. This record does not assert independent reproduction. ### USSM + Cause-Effect Knowledge Base Accuracy: 55. Result ID: 698cce008b517eb4331cdef8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"unsupervised","knowledge\_base":"Cause-Effect"}. This record does not assert independent reproduction. ### Single LM (partial) Accuracy: 53.3. Result ID: 698cce008b517eb4331cdefb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"unsupervised","variant":"partial"}. This record does not assert independent reproduction. ### USSM + Supervised Deepnet Accuracy: 53.3. Result ID: 698cce008b517eb4331cdf00. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"supervised"}. This record does not assert independent reproduction. ### Unsupervised Semantic Similarity Method (USSM) Accuracy: 48.3. Result ID: 698cce008b517eb4331cdef7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"supervision":"unsupervised"}. This record does not assert independent reproduction. ### Nicos Issak Accuracy: 48.3. Result ID: 698cce008b517eb4331cdefe. [Reporting paper]() Recorded evaluation configuration: {"supervision":"supervised","competition":"WS Challenge 2016"}. This record does not assert independent reproduction. ### Patric Dhondt Accuracy: 45. Result ID: 698cce008b517eb4331cdefd. [Reporting paper]() Recorded evaluation configuration: {"supervision":"supervised","competition":"WS Challenge 2016"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.