Ayushnangia/compactionbench-lme-text-subset download history

Ayushnangia/compactionbench-lme-text-subset is a question answering dataset on the Hugging Face Hub. In the last 30 days it was downloaded 21 times (5 in the last 7 days), and 212 times in total. It ranks #361,218 among datasets by monthly downloads.

CompactionBench LME Text Subset 20 questions from LongMemEval-V2 filtered for text-answerable content. Each task has ~160k tokens of web agent trajectory context (thoughts, actions, page states). All answers are confirmed present in the text context. Used for testing context compaction in

Open Ayushnangia/compactionbench-lme-text-subset on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.