Summarization on CLPsych Task 3.1 2026 (test)
86.6CSAurevia
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| AureviaSub. ID=6934542026.05 | 86.6 | 62.5 | 18.5 | 22.6 | 6.25 | |
| psytechlabSub. ID=6899732026.05 | 85.7 | 57.1 | 7.8 | 14.7 | 7.25 | |
| MERONYM_LABSSub. ID=6942292026.05 | 80.1 | 65.9 | 26.6 | 34.5 | 4 | |
| JNLPSub. ID=6913152026.05 | 79.1 | 66.6 | 11.7 | 16.4 | 8 | |
| CUNYSub. ID=6942162026.05 | 78.9 | 71.4 | 29.2 | 29.5 | 5.5 | |
| McMaster NLPSub. ID=6941892026.05 | 77 | 76.1 | 20.8 | 25.5 | 8 | |
| DreamerNLplusSub. ID=6939642026.05 | 73.5 | 76.7 | 28.5 | 34.5 | 5.25 | |
| NoviceTrioSub. ID=6939132026.05 | 70.5 | 77.1 | 31.8 | 34.1 | 5.75 | |
| CSE_IIT_RoparSub. ID=6943112026.05 | 68.8 | 81.2 | 24.2 | 30.6 | 8.5 | |
| USAISub. ID=6939122026.05 | 68.1 | 84.9 | 33.3 | 36.5 | 6.25 | |
| MKCSub. ID=6877772026.05 | 65.4 | 83.4 | 28.4 | 35.9 | 7 | |
| CtbuYSub. ID=6914462026.05 | 61.5 | 84.8 | 23.2 | 31.7 | 10 | |
| ULLSub. ID=6946692026.05 | 58.5 | 84.6 | 26.2 | 32 | 9.25 | |
| Submission 4*Backbone=gpt, Prompting Strategy=Label-augmented ICL variant, Data Augmentation=predicted labels from Tasks 1.1 and 2, Status=made during the analysis phase2026.05 | 0.818 | 0.621 | 0.305 | — | — | |
| Submission 1Backbone=gpt, Prompting Strategy=plain ICL, k=22026.05 | 0.797 | 0.696 | 0.283 | 0.249 | — | |
| Submission 3Backbone=gpt, Prompting Strategy=Label-augmented ICL variant, Data Augmentation=gold subelements and change labels2026.05 | 0.789 | 0.714 | 0.292 | 0.295 | — | |
| Official baseline 2Strategy=pipeline approach, Context Enrichment=predictions from top performing baselines from Task 1 and 22026.05 | 0.767 | 0.745 | 0.269 | 0.235 | — | |
| Official baseline 1Backbone=Llama-3.1-8B, Strategy=zero-shot2026.05 | 0.763 | 0.753 | 0.255 | 0.226 | — | |
| Submission 2Backbone=gpt, Prompting Strategy=summary-of-summaries approach2026.05 | 0.722 | 0.808 | 0.244 | 0.218 | — |