Unverified50% confidenceFactExact time
In LExGraph, hybrid retrieval is responsible for locating seeds among memory nodes, and neighbor expansion supplements this with structural associations, forming a two-stage evidence acquisition process of "locate first, then diffuse".
1
Sources
50%
Confidence
Long-term
Relevance
11/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe framework is further extended to episodic confounded Markov decision processes, where under structural and completeness conditions, complementary source data can identify the reward and transition mechanisms of the target domain.65% similarUnverifiedThe knowledge generalization mechanism enables Procedural Graphs to identify common patterns across tasks and reuse optimized graph structure fragments to achieve knowledge transfer.64% similarUnverified自注意力机制允许模型在单次计算中直接建立序列中任意两个位置元素之间的语义依赖关系，解决了长距离依赖问题并支持并行计算64% similarUnverifiedDiffusionGemma依赖双向注意力，让前后位置互相照应，这是其生成质量不输自回归的核心机制64% similarUnverifiedResearch from the MBZUAI NLP team proposes the graph-guided benchmark LExMem and the graph-based memory framework LExGraph for memory modeling and evaluation in long-horizon exploration tasks.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1138197API
curl https://kongchang.com/api/v1/knowledge/claims/1138197MCP
get_claim(id=1138197)