marcodsn/altered-riddles
Altered Riddles: A Benchmark for Conditioned Override in LLMs Altered Riddles is a benchmark designed to measure Conditioned Override Rate (COR) — the tendency of language models to fall back on memorized answers when a familiar riddle is subtly modified. This benchmark helps identify when models solve problems based on memorization versus reasoning. Dataset Description The benchmark measures how often models fail to adapt when a well-known riddle is altered in… See the full description on the dataset page: https://huggingface.co/datasets/marcodsn/altered-riddles.
| Author / organization | marcodsn |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 358 (as reported by Hugging Face) |
| Likes | 11 |
| Created | 3 May 2025 |
| Last updated at source | 18 April 2026 |
Popularity, last 2 days
358 now (-5 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 17 Oct 2026 Description: Altered Riddles: A Benchmark for Conditioned Override in LLMs Altered Riddles is a benchmark designed to measure Conditioned Override Rate (COR) — the tendency of language models to fall back on memorized answers when a familiar riddle is s → Altered Riddles: A Benchmark for Conditioned Override in LLMs Altered Riddles is a benchmark designed to measure Conditioned Override Rate (COR) — the tendency of language models to fall back on memorized answers when a familiar riddle is s
Changes detected by Silytics between refreshes of the source.
More from marcodsn
- marcodsn/academic-chainsDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.