Papr scores 96.0% on LongMemEval with schema-aware retrieval for long-term memory

Papr answered 480 of 500 LongMemEval questions correctly (96.0%), the highest published score it has found, using the benchmark's own answer prompt and grader. Its 95.4% configuration sends the answer model 16.5k tokens per question, about a third of what Hindsight sends, and with gpt-5-mini it reaches 94.4%. Retrieval places all evidence sessions in the top 5 for 97.9% of answerable questions.