Series
Memory systems
A numbered series on memory systems.
- 01
We retrieved the memory, then dropped the answer
A 200-question conflict run scored 0.800 at the parent-Memory boundary and 0.635 at the evidence boundary. The gap was in what reached the reader.
- 02
How we took LongMemEval from 80.0 to 94.2 without touching retrieval
LongMemEval-S climbed from 80.0 to 94.2 while retrieval recall@10 stayed at 0.99. The gains came from evidence budget, reader generation, and instructions.
- 03
Why we do not extract facts by default
A 500-question memory experiment cut reader input tenfold but lost 6.4 points of answer accuracy. The missing details were gone before retrieval.