We lifted 3 functions out of this paper's own repositories and ran 3 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 |
|---|---|---|
| yaojh18/Varying-Shades-of-Wrong | canonical | 3 of 3 |
| Function | Status | Where it lives |
|---|---|---|
| calculate_ece | Ran | yaojh18/Varying-Shades-of-Wrong/preference_optimization/evaluate.py pointer only (licence: NONE) · get_code("bc70bd9f9c7c1913") |
| calculate_metrics | Ran | yaojh18/Varying-Shades-of-Wrong/preference_optimization/evaluate.py pointer only (licence: NONE) · get_code("29adfabf77b2b461") |
| find_longest_path | Ran | yaojh18/Varying-Shades-of-Wrong/preference_optimization/evaluate.py pointer only (licence: NONE) · get_code("ea8ae4a67865c068") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
In the absence of abundant reliable annotations for challenging tasks and contexts, how can we expand the frontier of LLM capabilities with potentially wrong answers? We focus on two research questions: (1) Can LLMs generate reliable preferences among wrong options? And if so, (2) Would alignment with such wrong-over-wrong preferences be helpful? We employ methods based on self-consistency, token probabilities, and LLM-as-a-judge to elicit wrong-over-wrong preferences, and fine-tune language models with preference optimization approaches using these synthesized preferences. Extensive experiments with seven LLMs and eight datasets demonstrate that (1) LLMs do have preliminary capability in distinguishing various shades of wrong, achieving up to 20.9% higher performance than random guess; (2) Alignment with wrong-over-wrong preferences helps LLMs to produce less wrong and sometimes even outright correct answers, while overall improving model calibration.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2410.11055")
get_code_for_paper("2410.11055")
have("2410.11055")
Connect an agent — have() is free.