{"slug": "speakermem-r1-speaker-centered-dual-track-memory-for-multi-party-dialogue", "title": "SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue", "summary": "Researchers submitted SpeakerMem-R1, a speaker-centered dual-track memory system for multi-party dialogue, to arXiv on 22 Sep 2026. SpeakerMem-R1 achieved 62.33% on the publicly reported EverMemBench leaderboard from EverMind-AI, the best reported result among the latest state-of-the-art frameworks, and 70.85% on all 1,986 LoCoMo questions. The system stores speaker-labeled verbatim messages and derived states in person-level and group-level views, and its Writer-R1 was trained with SpeakerLevenshtein and speaker-conditioned GRPO, raising the SFT Writer's mean accuracy from 57.38% to 68.20% in a controlled evaluation of 305 questions.", "body_md": "# Computer Science > Computation and Language\n\n  [Submitted on 22 Sep 2026]\n\n# Title:SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue\n\n[View PDF](http://arxiv.org/pdf/2609.26780v1)\n\n[HTML (experimental)](https://arxiv.org/html/2609.26780v1)\n\nAbstract:Long-term conversational memory in multi-party settings requires more than retrieving relevant content from long-term conversations: it must distinguish who said what, whom each statement concerns, how individuals perceive one another, what information is shared by the group, and how states change over time. Recent studies on multi-party dialogue benchmarks show that existing general-purpose LLM memory systems tend to lose person and group relations or struggle to integrate clues distributed across members, groups, and time. Together, these issues reveal two core bottlenecks: message attribution and relational understanding in multi-party dialogue, and state reconstruction from interleaved histories. To address both, we propose $\\textbf{SpeakerMem-R1}$: its dual-track memory stores speaker-labeled verbatim messages and derived states organized into person-level and group-level views, then combines evidence from both tracks by entity, event, and time at query time. To reduce attribution and update errors during structured memory construction while enabling local deployment, we train Writer-R1 with SpeakerLevenshtein and speaker-conditioned GRPO. On GroupMemBench, SocialMemBench, and EverMemBench, SpeakerMem-R1 achieves binary accuracies of 47.9%, 69.2%, and 61.9%, respectively. On the publicly reported EverMemBench leaderboard from EverMind-AI, we achieves 62.33%, the best reported result among the latest state-of-the-art frameworks. It also achieves 70.85% on all 1,986 LoCoMo questions, which we use as a two-person long-term conversation boundary test. In a controlled evaluation of 305 questions, RL raises the SFT Writer's mean accuracy from 57.38% to 68.20%. We report both binary accuracy and token-F1, and ablations show that the verbatim and structured tracks, as well as person-level and group-level views, are complementary under the standardized evaluation interface.\n    \n\n### Current browse context:\n\ncs.CL\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/speakermem-r1-speaker-centered-dual-track-memory-for-multi-party-dialogue", "canonical_source": "http://arxiv.org/abs/2609.26780v1", "published_at": "2026-09-23 14:10:07+00:00", "updated_at": "2026-09-23 15:31:14.758036+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "natural-language-processing", "ai-research", "machine-learning"], "entities": ["SpeakerMem-R1", "Writer-R1", "GroupMemBench", "SocialMemBench", "EverMemBench", "EverMind-AI", "LoCoMo", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/speakermem-r1-speaker-centered-dual-track-memory-for-multi-party-dialogue", "markdown": "https://wpnews.pro/news/speakermem-r1-speaker-centered-dual-track-memory-for-multi-party-dialogue.md", "text": "https://wpnews.pro/news/speakermem-r1-speaker-centered-dual-track-memory-for-multi-party-dialogue.txt", "jsonld": "https://wpnews.pro/news/speakermem-r1-speaker-centered-dual-track-memory-for-multi-party-dialogue.jsonld"}}