LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K.
CoRR(2024)
AI Read Science
Must-Reading Tree
Example

Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined