Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.11628 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908961405403136 |
|---|---|
| author | Wu, Yuqian Chen, Wei Huang, Zhengjun Chen, Junle Liu, Qingxiang Wang, Kai Zhou, Xiaofang Liang, Yuxuan |
| author_facet | Wu, Yuqian Chen, Wei Huang, Zhengjun Chen, Junle Liu, Qingxiang Wang, Kai Zhou, Xiaofang Liang, Yuxuan |
| contents | Existing conversational memory systems rely on complex hierarchical summarization or reinforcement learning to manage long-term dialogue history, yet remain vulnerable to context dilution as conversations grow. In this work, we offer a different perspective: the primary bottleneck may lie not in memory architecture, but in the \textit{Signal Sparsity Effect} within the latent knowledge manifold. Through controlled experiments, we identify two key phenomena: \textit{Decisive Evidence Sparsity}, where relevant signals become increasingly isolated with longer sessions, leading to sharp degradation in aggregation-based methods; and \textit{Dual-Level Redundancy}, where both inter-session interference and intra-session conversational filler introduce large amounts of non-informative content, hindering effective generation. Motivated by these insights, we propose \method, a minimalist framework that brings conversational memory back to basics, relying solely on retrieval and generation via Turn Isolation Retrieval (TIR) and Query-Driven Pruning (QDP). TIR replaces global aggregation with a max-activation strategy to capture turn-level signals, while QDP removes redundant sessions and conversational filler to construct a compact, high-density evidence set. Extensive experiments on multiple benchmarks demonstrate that \method achieves robust performance across diverse settings, consistently outperforming strong baselines while maintaining high efficiency in tokens and latency, establishing a new minimalist baseline for conversational memory. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_11628 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Back to Basics: Let Conversational Agents Remember with Just Retrieval and Generation Wu, Yuqian Chen, Wei Huang, Zhengjun Chen, Junle Liu, Qingxiang Wang, Kai Zhou, Xiaofang Liang, Yuxuan Computation and Language Artificial Intelligence Existing conversational memory systems rely on complex hierarchical summarization or reinforcement learning to manage long-term dialogue history, yet remain vulnerable to context dilution as conversations grow. In this work, we offer a different perspective: the primary bottleneck may lie not in memory architecture, but in the \textit{Signal Sparsity Effect} within the latent knowledge manifold. Through controlled experiments, we identify two key phenomena: \textit{Decisive Evidence Sparsity}, where relevant signals become increasingly isolated with longer sessions, leading to sharp degradation in aggregation-based methods; and \textit{Dual-Level Redundancy}, where both inter-session interference and intra-session conversational filler introduce large amounts of non-informative content, hindering effective generation. Motivated by these insights, we propose \method, a minimalist framework that brings conversational memory back to basics, relying solely on retrieval and generation via Turn Isolation Retrieval (TIR) and Query-Driven Pruning (QDP). TIR replaces global aggregation with a max-activation strategy to capture turn-level signals, while QDP removes redundant sessions and conversational filler to construct a compact, high-density evidence set. Extensive experiments on multiple benchmarks demonstrate that \method achieves robust performance across diverse settings, consistently outperforming strong baselines while maintaining high efficiency in tokens and latency, establishing a new minimalist baseline for conversational memory. |
| title | Back to Basics: Let Conversational Agents Remember with Just Retrieval and Generation |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2604.11628 |