LoCoMo

The 'LOCOMO' dataset/benchmark is used to evaluate long-term conversation capabilities of personalized AI agents by providing a structured framework for assessing their memory and reasoning abilities over extended interactions.

Papers using LoCoMo (33)

EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments2026Rethinking Memory as Continuously Evolving Connectivity2026Learning Query-Aware Budget-Tier Routing for Runtime Agent Memory2026Omni-simplemem: Autoresearch-guided Discovery Of Lifelong Multimodal Agent Memory2026AtomMem: Building Simple and Effective Memory System for LLM Agents via Atomic Facts2026A-TMA: Decoupling State-Aware Memory Failures in Long-Term Agent Memory2026MemPro: Agentic Memory Systems as Evolvable Programs2026DMF: A Deterministic Memory Framework for Conversational AI Agents2026Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents2026Learning What to Remember: Observability-Safe Memory Retention via Constrained Optimization for Long-Horizon Language Agents2026Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents2026User as Code: Executable Memory for Personalized Agents2026Mandol: An Agglomerative Agent Memory System for Long-Term Conversations2026Parallel Context Compaction for Long-Horizon LLM Agent Serving2026MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing2026MGRetrieval: Memory-Guided Reflective Retrieval for Long-Term Dialogue Agents2026MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents2026SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs2026Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents2025 · 1 citesScaling Self-Evolving Agents via Parametric Memory2026PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents2026AgentIR: A Workload-Adaptive Cascade Retrieval Substrate for Long-Term Conversational Memory2026Mitigating Provenance-Role Collapse in Long-Term Agents via Typed Memory Representation2026TReMu: Towards Neuro-Symbolic Temporal Reasoning for LLM-Agents with Memory in Multi-Session Dialogues2025 · 1 citesRethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory2026MemORAI: Memory Organization and Retrieval via Adaptive Graph Intelligence for LLM Conversational Agents2026Memrouter: Memory-as-embedding Routing For Long-term Conversational Agents2026All-Mem: Agentic Lifelong Memory via Dynamic Topology Evolution2026MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents2026Memora: A Harmonic Memory Representation Balancing Abstraction and Specificity2026Membox: Weaving Topic Continuity into Long-Range Memory for LLM Agents2026PISA: A Pragmatic Psych-Inspired Unified Memory System for Enhanced AI Agency2025RGMem: Renormalization Group-inspired Memory Evolution for Language Agents2025