LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
Paper
• 2410.10813 • Published
• 14
custom_id stringlengths 8 17 | question stringlengths 33 243 | answer stringlengths 1 390 | full_input stringlengths 488k 536k | full_input_tokens int64 109k 115k | focused_input stringlengths 278 4k | focused_input_tokens int64 67 874 |
|---|---|---|---|---|---|---|
0a995998 | How many items of clothing do I need to pick up or return from a store? | 3 | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 113,933 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 320 |
b5ef892d | How many days did I spend on camping trips in the United States this year? | 8 days. | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 112,807 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 247 |
e831120c | "How many weeks did it take me to watch all the Marvel Cinematic Universe movies and the main Star W(...TRUNCATED) | 3.5 weeks | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 112,186 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 234 |
3a704032 | How many plants did I acquire in the last month? | 3 | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 113,919 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 213 |
aae3761f | How many hours in total did I spend driving to my three road trip destinations combined? | 15 hours for getting to the three destinations (or 30 hours for the round trip) | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 111,512 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 331 |
gpt4_f2262a51 | How many different doctors did I visit? | "I visited three different doctors: a primary care physician, an ENT specialist, and a dermatologist(...TRUNCATED) | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 111,685 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 506 |
dd2973ad | What time did I go to bed on the day before I had a doctor's appointment? | 2 AM | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 113,393 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 244 |
c4a1ceb8 | How many different types of citrus fruits have I used in my cocktail recipes? | 3 | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 112,253 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 435 |
gpt4_a56e767c | How many movie festivals that I attended? | I attended four movie festivals. | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 112,933 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 740 |
6cb6f249 | How many days did I take social media breaks in total? | 17 days | "I will give you several history chats between you and a user. Please answer the question based on t(...TRUNCATED) | 111,471 | "I will give you several user history chats. Please answer the question based on the relevant chat h(...TRUNCATED) | 244 |
End of preview. Expand
in Data Studio
Cleaned Version of LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory
This dataset is a cleaned version of LongMemEval by Wu et al. (2024).
This dataset was used in Context Rot.
Modifications
- Removed ambiguous question-answers
- Fixed focused (orcale) version to ensure question can be fully answered by input
Citation
If you use this dataset, please cite the original authors:
@article{wu2024longmemeval, title={LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory}, author={Di Wu and Hongwei Wang and Wenhao Yu and Yuwei Zhang and Kai-Wei Chang and Dong Yu}, year={2024}, eprint={2410.10813}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2410.10813}, }
- Downloads last month
- 75
Size of downloaded dataset files:
158 MB
Size of the auto-converted Parquet files:
75.8 MB
Number of rows:
306