We lifted 13 functions out of this paper's own repositories and ran 0 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 |
|---|---|---|
| jordantab/llm-robustness-experiment | canonical | 0 of 13 |
| Function | Status | Where it lives |
|---|---|---|
| calculate_grouped_metrics | Not yet run | jordantab/llm-robustness-experiment/utilities/calculate_metrics.py pointer only (licence: NONE) · get_code("a1b8498b1a2f8c75") |
| calculate_metrics | Not yet run | jordantab/llm-robustness-experiment/utilities/calculate_pb_baseline.py pointer only (licence: NONE) · get_code("54807fe1fc74ab13") |
| calculate_metrics | Not yet run | jordantab/llm-robustness-experiment/utilities/parse.py pointer only (licence: NONE) · get_code("c8436ea20f798b9a") |
| do_pearson | Not yet run | jordantab/llm-robustness-experiment/utilities/correlation.py pointer only (licence: NONE) · get_code("7eb7968dc3063055") |
| evaluate_advplusplus_sst2 | Not yet run | jordantab/llm-robustness-experiment/utilities/evaluated_everything.py pointer only (licence: NONE) · get_code("a61897c5c3886302") |
| evaluate_advplusplus_task | Not yet run | jordantab/llm-robustness-experiment/utilities/evaluated_everything.py pointer only (licence: NONE) · get_code("56fe6b8f7f36c182") |
| evaluate_flipkart | Not yet run | jordantab/llm-robustness-experiment/utilities/evaluated_everything.py pointer only (licence: NONE) · get_code("8810883cafa51f67") |
| extract_prompts_from_file | Not yet run | jordantab/llm-robustness-experiment/utilities/extract_prompt.py pointer only (licence: NONE) · get_code("78e27148f26a063d") |
| extract_prompts_from_file | Not yet run | jordantab/llm-robustness-experiment/utilities/extract_prompt_quota.py pointer only (licence: NONE) · get_code("1f35d03d71695e79") |
| get_metric_pairs | Not yet run | jordantab/llm-robustness-experiment/utilities/correlation.py pointer only (licence: NONE) · get_code("20bb3b3f013e2d28") |
| get_structure | Not yet run | jordantab/llm-robustness-experiment/utilities/json_structure.py pointer only (licence: NONE) · get_code("600ae85bfd3e7930") |
| lin_fit | Not yet run | jordantab/llm-robustness-experiment/utilities/correlation.py pointer only (licence: NONE) · get_code("9b0a0f8837543b40") |
| parse_progress_files | Not yet run | jordantab/llm-robustness-experiment/utilities/parse.py pointer only (licence: NONE) · get_code("5112128f4d45c216") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
The increasing reliance on large language models (LLMs) for diverse applications necessitates a thorough understanding of their robustness to adversarial perturbations and out-of-distribution (OOD) inputs. In this study, we investigate the correlation between adversarial robustness and OOD robustness in LLMs, addressing a critical gap in robustness evaluation. By applying methods originally designed to improve one robustness type across both contexts, we analyze their performance on adversarial and out-of-distribution benchmark datasets. The input of the model consists of text samples, with the output prediction evaluated in terms of accuracy, precision, recall, and F1 scores in various natural language inference tasks. Our findings highlight nuanced interactions between adversarial robustness and OOD robustness, with results indicating limited transferability between the two robustness types. Through targeted ablations, we evaluate how these correlations evolve with different model sizes and architectures, uncovering model-specific trends: smaller models like LLaMA2-7b exhibit neutral correlations, larger models like LLaMA2-13b show negative correlations, and Mixtral demonstrates positive correlations, potentially due to domain-specific alignment. These results underscore the importance of hybrid robustness frameworks that integrate adversarial and OOD strategies tailored to specific models and domains. Further research is needed to evaluate these interactions across larger models and varied architectures, offering a pathway to more reliable and generalizable LLMs.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2412.10535")
get_code_for_paper("2412.10535")
have("2412.10535")
Connect an agent — have() is free.