We lifted 5 functions out of this paper's own repositories and ran 5 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 |
|---|---|---|
| stefanhgm/patient_summaries_with_llms | canonical | 5 of 5 |
| Function | Status | Where it lives |
|---|---|---|
| create_icl_example_v1 | Ran | stefanhgm/patient_summaries_with_llms/gpt-4/run_hallucination_detection.py code served (permissive licence) · get_code("1dfa22c583d35631") |
| extract_di | Ran | stefanhgm/patient_summaries_with_llms/preprocess/process_mimic_summaries.py code served (permissive licence) · get_code("34f40108fd90e5f1") |
| extract_hc | Ran | stefanhgm/patient_summaries_with_llms/preprocess/process_mimic_summaries.py code served (permissive licence) · get_code("d6b6f331b351e1a3") |
| parse_hallucination_detection_results | Ran | stefanhgm/patient_summaries_with_llms/gpt-4/run_hallucination_detection.py code served (permissive licence) · get_code("c40d85bacef9a266") |
| remove_empty_and_short_summaries | Ran | stefanhgm/patient_summaries_with_llms/preprocess/process_mimic_summaries.py code served (permissive licence) · get_code("115edd746212499f") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Patients often face difficulties in understanding their hospitalizations, while healthcare workers have limited resources to provide explanations. In this work, we investigate the potential of large language models to generate patient summaries based on doctors' notes and study the effect of training data on the faithfulness and quality of the generated summaries. To this end, we release (i) a rigorous labeling protocol for errors in medical texts and (ii) a publicly available dataset of annotated hallucinations in 100 doctor-written and 100 generated summaries. We show that fine-tuning on hallucination-free data effectively reduces hallucinations from 2.60 to 1.55 per summary for Llama 2, while preserving relevant information. We observe a similar effect on GPT-4 (0.70 to 0.40), when the few-shot examples are hallucination-free. We also conduct a qualitative evaluation using hallucination-free and improved training data. We find that common quantitative metrics do not correlate well with faithfulness and quality. Finally, we test GPT-4 for automatic hallucination detection, which clearly outperforms common baselines.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2402.15422")
get_code_for_paper("2402.15422")
have("2402.15422")
Connect an agent — have() is free.