# PairPref: When Should Memory Guide the Answer? A Benchmark for Contextual Preference Use

arXiv: https://arxiv.org/abs/2609.34526 · HTML: https://mlsanity.com/paper/2609.34526/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 46 of its 55 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 30 of 46)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory](https://mlsanity.com/paper/2504.19413/) (2025)
- [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://mlsanity.com/paper/2410.10813/) (2024)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [MemoryBank: Enhancing Large Language Models with Long-Term Memory](https://mlsanity.com/paper/2305.10250/) (2023)
- [LaMP: When Large Language Models Meet Personalization](https://mlsanity.com/paper/2304.11406/) (2023)
- [Zep: A Temporal Knowledge Graph Architecture for Agent Memory](https://mlsanity.com/paper/2501.13956/) (2025)
- [Memory in the Age of AI Agents](https://mlsanity.com/paper/2512.13564/) (2025)
- [HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models](https://mlsanity.com/paper/2405.14831/) (2024)
- [Beyond Goldfish Memory: Long-Term Open-Domain Conversation](https://mlsanity.com/paper/2107.07567/) (2021)
- [Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning](https://mlsanity.com/paper/2508.19828/) (2025)
- [From RAG to Memory: Non-Parametric Continual Learning for Large Language Models](https://mlsanity.com/paper/2502.14802/) (2025)
- [LightMem: Lightweight and Efficient Memory-Augmented Generation](https://mlsanity.com/paper/2510.18866/) (2025)
- [Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and Personalized Responses at Scale](https://mlsanity.com/paper/2504.14225/) (2025)
- [Memory OS of AI Agent](https://mlsanity.com/paper/2506.06326/) (2025)
- [Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs](https://mlsanity.com/paper/2502.09597/) (2025)
- [LongLaMP: A Benchmark for Personalized Long-form Text Generation](https://mlsanity.com/paper/2407.11016/) (2024)
- [HaluMem: Evaluating Hallucinations in Memory Systems of Agents](https://mlsanity.com/paper/2511.03506/) (2025)

## Abstract (arXiv)

Memory-augmented assistants use retrieved preferences to guide their responses. A small change in the situation can change whether a preference is appropriate while barely affecting its retrieval similarity. Memory benchmarks typically test whether systems store and retrieve preferences, with less attention to when those preferences should apply. We introduce PairPref, a benchmark of contextual preference use. Each pair changes only the situation, keeping the preference, request, and four candidate replies fixed. The preference remains valid in both situations. In the selection track, models must choose the reply that applies the preference only where appropriate. In the free-generation track, they must decide when to apply it without seeing candidate replies. Both tracks use the same 1,227 pairs across 45 preferences and eight situation categories. We evaluate eight models, most of which achieve selection scores (Delta) of 51 to 65 points. In free generation, however, both responses are appropriate for their respective situations in only 3.6% to 18.3% of pairs. Models continue to apply the preference in both situations even with fewer retrieved memories, alternative presentation formats, and a stricter prompt. These results show that models still struggle to judge when user preferences apply and respond accordingly.