PersistBench: When Should Long-Term Memories Be Forgotten by LLMs?
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912866827763712 |
|---|---|
| author | Pulipaka, Sidharth Chen, Oliver Sharma, Manas Bajwa, Taaha S Raina, Vyas Sheth, Ivaxi |
| author_facet | Pulipaka, Sidharth Chen, Oliver Sharma, Manas Bajwa, Taaha S Raina, Vyas Sheth, Ivaxi |
| contents | Conversational assistants are increasingly integrating long-term memory with large language models (LLMs). This persistence of memories, e.g., the user is vegetarian, can enhance personalization in future conversations. However, the same persistence can also introduce safety risks that have been largely overlooked. Hence, we introduce PersistBench to measure the extent of these safety risks. We identify two long-term memory-specific risks: cross-domain leakage, where LLMs inappropriately inject context from the long-term memories; and memory-induced sycophancy, where stored long-term memories insidiously reinforce user biases. We evaluate 18 frontier and open-source LLMs on our benchmark. Our results reveal a surprisingly high failure rate across these LLMs - a median failure rate of 53% on cross-domain samples and 97% on sycophancy samples. To address this, our benchmark encourages the development of more robust and safer long-term memory usage in frontier conversational systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_01146 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PersistBench: When Should Long-Term Memories Be Forgotten by LLMs? Pulipaka, Sidharth Chen, Oliver Sharma, Manas Bajwa, Taaha S Raina, Vyas Sheth, Ivaxi Artificial Intelligence I.2.7 Conversational assistants are increasingly integrating long-term memory with large language models (LLMs). This persistence of memories, e.g., the user is vegetarian, can enhance personalization in future conversations. However, the same persistence can also introduce safety risks that have been largely overlooked. Hence, we introduce PersistBench to measure the extent of these safety risks. We identify two long-term memory-specific risks: cross-domain leakage, where LLMs inappropriately inject context from the long-term memories; and memory-induced sycophancy, where stored long-term memories insidiously reinforce user biases. We evaluate 18 frontier and open-source LLMs on our benchmark. Our results reveal a surprisingly high failure rate across these LLMs - a median failure rate of 53% on cross-domain samples and 97% on sycophancy samples. To address this, our benchmark encourages the development of more robust and safer long-term memory usage in frontier conversational systems. |
| title | PersistBench: When Should Long-Term Memories Be Forgotten by LLMs? |
| topic | Artificial Intelligence I.2.7 |
| url | https://arxiv.org/abs/2602.01146 |