Survey Generation on ACL Survey Dataset (test)
3.725Logical Structure ScoreSTRUCTSURVEY
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| STRUCTSURVEYLLM-as-a-Judge=claude-sonnet-4-5-202509292026.05 | 3.725 | 2.808 | 3.025 | 3.767 | — | |
| SURVEYFORGELLM-as-a-Judge=claude-sonnet-4-5-202509292026.05 | 3.575 | 2.667 | 3.042 | 3.65 | — |