# AMR Parsing on LDC2014T12 (Full) Source: [AMR Parsing on LDC2014T12 (Full)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:42:58.849Z Evaluation of Abstract Meaning Representation (AMR) parsing performance on the full set of the LDC2014T12 dataset, measured by F1 score. 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 | AMR parsing | | Dataset (including any stated split) | LDC2014T12 (Full) | | Metric | F1 Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 32 of 32 | | 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 - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | F1 Score | Result date | Report paper | | --- | --- | --- | --- | | Goodman et al. (2016) | 70 | Not recorded | 68fa638a1e209131ce9ff67e | | Transition-based AMR Parser | 68.4 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 68.1 | Not recorded | 68fa64e21e209131cea070a1 | | Wang and Xue | 68.1 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 67.7 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 67.4 | Not recorded | 68fa64e21e209131cea070a1 | | Pust et al. | 67.1 | Not recorded | 68fa64e21e209131cea070a1 | | Flanigan et al. (2016) | 66 | Not recorded | 68fa638a1e209131ce9ff67e | | Zhou et al. (2016) | 66 | Not recorded | 68fa638a1e209131ce9ff67e | | Pust et al. (2015) | 66 | Not recorded | 68fa638a1e209131ce9ff67e | | Wang et al. (2015a) | 66 | Not recorded | 68fa638a1e209131ce9ff67e | | Damonte et al. | 66 | Not recorded | 68fa64e21e209131cea070a1 | | Wang et al. | 66 | Not recorded | 68fa64e21e209131cea070a1 | | Zhou et al. | 66 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 65.2 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 64.7 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 64.6 | Not recorded | 68fa64e21e209131cea070a1 | | Damonte et al. (2016) | 64 | Not recorded | 68fa638a1e209131ce9ff67e | | Stack-LSTM AMR parser | 64 | Not recorded | 68fa638a1e209131ce9ff67e | | Ballesteros and Al-Onaizan | 64 | Not recorded | 68fa64e21e209131cea070a1 | | Transition-based AMR Parser | 63.9 | Not recorded | 68fa64e21e209131cea070a1 | | Stack-LSTM AMR parser | 63 | Not recorded | 68fa638a1e209131ce9ff67e | | Stack-LSTM AMR parser | 63 | Not recorded | 68fa638a1e209131ce9ff67e | | Ballesteros and Al-Onaizan | 63 | Not recorded | 68fa64e21e209131cea070a1 | | Ballesteros and Al-Onaizan | 63 | Not recorded | 68fa64e21e209131cea070a1 | | Stack-LSTM AMR parser | 62 | Not recorded | 68fa638a1e209131ce9ff67e | | Damonte et al. (2016) | 61 | Not recorded | 68fa638a1e209131ce9ff67e | | Pust et al. (2015) | 61 | Not recorded | 68fa638a1e209131ce9ff67e | | Stack-LSTM AMR parser | 61 | Not recorded | 68fa638a1e209131ce9ff67e | | Wang et al. (2015b) | 59 | Not recorded | 68fa638a1e209131ce9ff67e | | Stack-LSTM AMR parser | 59 | Not recorded | 68fa638a1e209131ce9ff67e | | Flanigan et al. (2014) | 58 | Not recorded | 68fa638a1e209131ce9ff67e | ### Goodman et al. (2016) F1 Score: 70. Result ID: 698bfa53f729bc7f68e730cd. [Reporting paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","Pipeline System":"true"}. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 68.4. Result ID: 698c910c0b614d12cff1a5c9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word + POS","Aligner":"Our aligner","Ensemble size":"10"}. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 68.1. Result ID: 698c910c0b614d12cff1a5c7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word only","Aligner":"Our aligner","Ensemble size":"10"}. This record does not assert independent reproduction. ### Wang and Xue F1 Score: 68.1. Result ID: 698c910c0b614d12cff1a5d1. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 67.7. Result ID: 698c910c0b614d12cff1a5c8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word + POS","Aligner":"Our aligner","Ensemble size":"3"}. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 67.4. Result ID: 698c910c0b614d12cff1a5c6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word only","Aligner":"Our aligner","Ensemble size":"3"}. This record does not assert independent reproduction. ### Pust et al. F1 Score: 67.1. Result ID: 698c910c0b614d12cff1a5cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Flanigan et al. (2016) F1 Score: 66. Result ID: 698bfa53f729bc7f68e730ca. [Reporting paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","Pipeline System":"true"}. This record does not assert independent reproduction. ### Zhou et al. (2016) F1 Score: 66. Result ID: 698bfa53f729bc7f68e730ce. [Reporting paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","SRL":"true","Pipeline System":"true"}. This record does not assert independent reproduction. ### Pust et al. (2015) F1 Score: 66. Result ID: 698bfa53f729bc7f68e730d0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LM":"true","NER":"true","WordNet":"true"}. This record does not assert independent reproduction. ### Wang et al. (2015a) F1 Score: 66. Result ID: 698bfa53f729bc7f68e730d2. [Reporting paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","SRL":"true","Pipeline System":"true"}. This record does not assert independent reproduction. ### Damonte et al. F1 Score: 66. Result ID: 698c910c0b614d12cff1a5cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Wang et al. F1 Score: 66. Result ID: 698c910c0b614d12cff1a5ce. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Zhou et al. F1 Score: 66. Result ID: 698c910c0b614d12cff1a5d0. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 65.2. Result ID: 698c910c0b614d12cff1a5c5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word + POS","Aligner":"Our aligner","Ensemble size":"1"}. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 64.7. Result ID: 698c910c0b614d12cff1a5c3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word only","Aligner":"Our aligner","Ensemble size":"1"}. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 64.6. Result ID: 698c910c0b614d12cff1a5c4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word + POS","Aligner":"JAMR aligner","Ensemble size":"1"}. This record does not assert independent reproduction. ### Damonte et al. (2016) F1 Score: 64. Result ID: 698bfa53f729bc7f68e730cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","SRL":"true","Note":"trained with LDC2014 training set"}. This record does not assert independent reproduction. ### Stack-LSTM AMR parser F1 Score: 64. Result ID: 698bfa53f729bc7f68e730d8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-trained word embeddings":"true","Character-based representations":"true","POS":"true","DEP":"true"}. This record does not assert independent reproduction. ### Ballesteros and Al-Onaizan F1 Score: 64. Result ID: 698c910c0b614d12cff1a5cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input features":"Word + POS + DEP"}. This record does not assert independent reproduction. ### Transition-based AMR Parser F1 Score: 63.9. Result ID: 698c910c0b614d12cff1a5c2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Input features":"Word only","Aligner":"JAMR aligner","Ensemble size":"1"}. This record does not assert independent reproduction. ### Stack-LSTM AMR parser F1 Score: 63. Result ID: 698bfa53f729bc7f68e730d6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-trained word embeddings":"true","Character-based representations":"true"}. This record does not assert independent reproduction. ### Stack-LSTM AMR parser F1 Score: 63. Result ID: 698bfa53f729bc7f68e730d7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-trained word embeddings":"true","Character-based representations":"true","POS":"true"}. This record does not assert independent reproduction. ### Ballesteros and Al-Onaizan F1 Score: 63. Result ID: 698c910c0b614d12cff1a5ca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input features":"Word only"}. This record does not assert independent reproduction. ### Ballesteros and Al-Onaizan F1 Score: 63. Result ID: 698c910c0b614d12cff1a5cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Input features":"Word + POS"}. This record does not assert independent reproduction. ### Stack-LSTM AMR parser F1 Score: 62. Result ID: 698bfa53f729bc7f68e730d5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-trained word embeddings":"true","Character-based representations":"false"}. This record does not assert independent reproduction. ### Damonte et al. (2016) F1 Score: 61. Result ID: 698bfa53f729bc7f68e730cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","SRL":"true","Note":"LDC2015 pretrained model evaluated on LDC2014"}. This record does not assert independent reproduction. ### Pust et al. (2015) F1 Score: 61. Result ID: 698bfa53f729bc7f68e730cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LM":"true","NER":"true"}. This record does not assert independent reproduction. ### Stack-LSTM AMR parser F1 Score: 61. Result ID: 698bfa53f729bc7f68e730d4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-trained word embeddings":"false","Character-based representations":"true"}. This record does not assert independent reproduction. ### Wang et al. (2015b) F1 Score: 59. Result ID: 698bfa53f729bc7f68e730d1. [Reporting paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","NER":"true","Pipeline System":"true"}. This record does not assert independent reproduction. ### Stack-LSTM AMR parser F1 Score: 59. Result ID: 698bfa53f729bc7f68e730d3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Pre-trained word embeddings":"false","Character-based representations":"false"}. This record does not assert independent reproduction. ### Flanigan et al. (2014) F1 Score: 58. Result ID: 698bfa53f729bc7f68e730c9. [Reporting paper]() Recorded evaluation configuration: {"POS":"true","DEP":"true","Pipeline System":"true"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.