Liu et al.
Event-centric memory with query-aware graph augmentation boosts long-term dialogue reasoning
QGMem organizes conversations into event-indexed atomic units and builds a local graph for relevant events, improving retrieval and reasoning over ultra-long contexts.
Chinese Tech
Yichen Liu · Chunfeng Yuan · Haowei Liu · Wenjuan Li · Zefeng Lin · Bing Li · +2 more
Chinese Academy of Sciences · University of Chinese Academy of Sciences · Alibaba Group · PeopleAI Inc. · Renmin University of China
Research Digest··3 min read
Liu et al.
Why this paper
From Chinese Academy of Sciences and 5 others
In one line
QGMem stores long dialogue histories as event-indexed memory units and builds query-specific local graphs, improving retrieval, multi-hop reasoning, and conflict resolution across six benchmarks.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§