[KongchangAI]
Benchmarkmemory-grounded QA

Memory-Anchored Question Answering

A task paradigm for evaluating whether a system can correctly use stored memories to generate answers. Responses must be traceable to specific entries in the memory store rather than to the model's pre-trained knowledge, making the paradigm particularly challenging in dynamic collaborative scenarios.

Timeline (last 90 days)

Sep 28

The research team constructed two new datasets oriented toward collaborative scenarios and based on memory-grounded QA to evaluate HiCoMER

Unverified50%
Sep 28

Memory-grounded QA is a task paradigm specifically designed to evaluate whether a system can correctly utilize stored memories to answer questions, where answers must be traceable to specific entries in the memory store rather than relying on pretrained knowledge in the model's parameters

Unverified50%

All Facts (2)

Source Articles