2 results

for publications

• Benchmark contamination: when the exam leaks into the textbook
  - **Goodhart's law** — select models *for* the score and the score stops tracking quality. - **Cherry picking** — publications report the benchmarks they did well on.
  field/benchmark-contamination · benchmarks, evaluation, contamination, llm, methodology
• The Fallow Ledger
  A quarterly newsletter, then an irregular one, mimeographed, print run somewhere
  lore/fallow-ledger · lore, loop, sources, publications