SodaMem: Evidence-Grounded Temporal Graph Memory for LLM Agents
Fengrong Wan, Chengcan Wu, Ningtao Lyu
cs.AI
Submitted: 2026-08-08
Updated: 2026-08-11
Code: https://github.com/SodaMem/SodaMem
License: http://creativecommons.org/licenses/by/4.0/
The gist: Large language model (LLM) agents that assist users over weeks of conversation must remember what is currently true, not merely what was once said.
Terminology
Abstract
Large language model (LLM) agents that assist users over weeks of conversation must remember what is currently true, not merely what was once said. Flat RAG diaries and Markdown logs optimize needle retrieval but under-serve currency, provenance, and ordered temporal reasoning (Maharana et al. 2024; Wu et al. 2024; Packer et al. 2023; Chhikara et al. 2025). We present SodaMem, an evidence-grounded temporal graph memory that (i) extracts typed FactEvents with mandatory provenance spans, (ii) persists mention time, occurrence time, and validity with SUPERSEDES/CONTRADICTS/UPDATES edges under hybrid lexical-dense indexing, and (iii) answers via a planner-reader loop that gathers citable evidence before composing a final response. On LongMemEval-S, our store-of-record configuration reaches 92.8% accuracy (464/500; best of N=3) at mean 0.00161/question (approximately 18.3k tokens; median 0.00111 / approximately 14.6k) with deepseek-v4-flash. We compile public systems with estimable API cost into a cost table and cost-accuracy map; under these estimates SodaMem sits near the accuracy frontier at Flash-tier spend and strictly dominates several higher-cost, lower-accuracy points. Accuracy uses the same Flash model as reader and judge (self-grading); costs exclude ingest/judge and cross-system comparisons are compiled estimates rather than a single-harness bake-off.Our code is available at https://github.com/SodaMem/SodaMem
Sources
- LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues
- MemGround: Long-Term Memory Evaluation Kit for Large Language Models in Gamified Scenarios
- STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?
- Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
- A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agents
- A-MEM: Agentic Memory for LLM Agents
- Zep: A Temporal Knowledge Graph Architecture for Agent Memory
- SimpleMem: Efficient Lifelong Memory for LLM Agents
- RaMem: Contextual Reinstatement for Long-term Agentic Memory
- HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models
- Profile-Graph Memory for LLM Agents: Implicit Cross-Entity Traversal through Narrative Profiles
- Reliable Post-Retrieval Assembly for Agent Memory: Separating Evidence Extraction from Policy Execution
- Beyond the Context Window: A Cost-Performance Analysis of Fact-Based Memory vs. Long-Context LLMs for Persistent Agents
- TiMem: Temporal-Hierarchical Memory Consolidation for Long-Horizon Conversational Agents
- MemOS: A Memory OS for AI System
- Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection