We lifted 16 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| dongping-chen/mixset | canonical | 6 of 16 |
| Function | Status | Where it lives |
|---|---|---|
| filter_test_data | Ran | dongping-chen/mixset/methods/utils.py pointer only (licence: NONE) · get_code("ea13cd1bc8c737e1") |
| get_ll | Ran | dongping-chen/mixset/methods/metric_based.py pointer only (licence: NONE) · get_code("f83e125c58abd645") |
| get_supervised_model_prediction_multi_classes | Ran | dongping-chen/mixset/methods/supervised.py pointer only (licence: NONE) · get_code("68c8b968f55907b7") |
| select_train_data | Ran | dongping-chen/mixset/methods/utils.py pointer only (licence: NONE) · get_code("416e7dabab64e070") |
| timeit | Ran | dongping-chen/mixset/methods/utils.py pointer only (licence: NONE) · get_code("58ba64be72bc1b9e") |
| tokenize_and_mask | Ran | dongping-chen/mixset/methods/detectgpt.py pointer only (licence: NONE) · get_code("639b00cb6787d6d0") |
| count_masks | Not yet run | dongping-chen/mixset/methods/detectgpt.py pointer only (licence: NONE) · get_code("23c3416429f69fee") |
| evaluate_model | Not yet run | dongping-chen/mixset/methods/radar.py pointer only (licence: NONE) · get_code("23449ffb5d7ab7b0") |
| evaluate_sentinel | Not yet run | dongping-chen/mixset/methods/sentinel.py pointer only (licence: NONE) · get_code("8e3d09aab30cff33") |
| get_lls | Not yet run | dongping-chen/mixset/methods/metric_based.py pointer only (licence: NONE) · get_code("c1d751eac084bac6") |
| get_rank | Not yet run | dongping-chen/mixset/methods/metric_based.py pointer only (licence: NONE) · get_code("ec539267c749a0cc") |
| get_supervised_model_prediction | Not yet run | dongping-chen/mixset/methods/supervised.py pointer only (licence: NONE) · get_code("ffed340b4b1b0302") |
| process_dataset | Not yet run | dongping-chen/mixset/dataset_loader.py pointer only (licence: NONE) · get_code("d6120a808c697802") |
| process_spaces | Not yet run | dongping-chen/mixset/dataset_loader.py pointer only (licence: NONE) · get_code("c5d3444c3ea71db9") |
| process_text_truthfulqa_adv | Not yet run | dongping-chen/mixset/dataset_loader.py pointer only (licence: NONE) · get_code("dd7573db30a9bd4e") |
| replace_masks | Not yet run | dongping-chen/mixset/methods/detectgpt.py pointer only (licence: NONE) · get_code("3b992344e4b4f98b") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
With the rapid development and widespread application of Large Language Models (LLMs), the use of Machine-Generated Text (MGT) has become increasingly common, bringing with it potential risks, especially in terms of quality and integrity in fields like news, education, and science. Current research mainly focuses on purely MGT detection without adequately addressing mixed scenarios, including AI-revised Human-Written Text (HWT) or human-revised MGT. To tackle this challenge, we define mixtext, a form of mixed text involving both AI and human-generated content. Then, we introduce MixSet, the first dataset dedicated to studying these mixtext scenarios. Leveraging MixSet, we executed comprehensive experiments to assess the efficacy of prevalent MGT detectors in handling mixtext situations, evaluating their performance in terms of effectiveness, robustness, and generalization. Our findings reveal that existing detectors struggle to identify mixtext, particularly in dealing with subtle modifications and style adaptability. This research underscores the urgent need for more fine-grain detectors tailored for mixtext, offering valuable insights for future research. Code and Models are available at https://github.com/Dongping-Chen/MixSet.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2401.05952")
get_code_for_paper("2401.05952")
have("2401.05952")
Connect an agent — have() is free.