Unverified50% confidenceFactExact time
The LExMem benchmark contains 224 dialogues and 2,018 question-answering instances, all derived from graph structures.
1
Sources
50%
Confidence
Long-term
Relevance
11/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe total token count for a single conversation is the sum of input (context, instructions, history) and output (the model's generated response)66% similarUnverifiedIn 16 sessions, 1,326 records contained strings suspected to be credentials.62% similarUnverifiedThe question-answer pairs in the LExMem benchmark are not manually written and then directly paired with context, but automatically derived from graph structures to ensure that questions depend on cross-branch and cross-time memory associations.60% similarUnverifiedIn the long-text question (a 14,000-character material summarization and data question), both products kept their summaries under 300 characters, covered all five parts, got all five data items correct, and got all page numbers correct.60% similarUnverifiedThe research team used 20 LLMs to generate 44,782 stories in four languages from textual and visual stimuli, involving a total of 88,919 named characters.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1138189API
curl https://kongchang.com/api/v1/knowledge/claims/1138189MCP
get_claim(id=1138189)