16 References

16.1 This repository

README.md, the operational overview
ROADMAP.md, the 28-item build plan with findings recorded per item
docs/agent-memory-synthesis.md, the conceptual landscape the design grew from
docs/claimgraph-handoff.md, design decisions with rationale and the alternatives weighed
docs/memory-systems-comparison.md, the July 2026 field comparison
docs/memagent-2026-review.md, claimgraph against the ICLR 2026 MemAgents workshop
docs/consuming-auto-memory.md, the ambient loop design note
.claude/skills/claimgraph/SKILL.md, the usage judgment

Packer et al., MemGPT: Towards LLMs as Operating Systems, 2023. arXiv:2310.08560
Rasmussen et al., Zep: A Temporal Knowledge Graph Architecture for Agent Memory, 2025. arXiv:2501.13956
Chhikara et al., Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory, 2025. arXiv:2504.19413
Xu et al., A-MEM: Agentic Memory for LLM Agents, NeurIPS 2025. arXiv:2502.12110
Lin et al., Sleep-time Compute, 2025. arXiv:2504.13171
engram, codebase experiential memory with conflict surfacing
ctxgraph, bi-temporal episode graph in SQLite
Graphiti; spec-kit; the OpenAI temporal-agents cookbook

Gloaguen, Mündler, Müller, Raychev, Vechev, Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents? (oral). arXiv:2602.11988
StructMemEval: Evaluating Memory Structure in LLM Agents (oral). arXiv:2602.11243
ALMA: Learning to Continually Learn via Meta-learning Agentic Memory Designs (oral). arXiv:2602.07755
AMA-Bench: Evaluating Long-Horizon Memory for Agentic Applications (oral). ICLR virtual 10021275
Diagnosing Retrieval vs. Utilization Bottlenecks in LLM Agent Memory. ICLR virtual 10021251
TierMem: From Lossy to Verified, a Provenance-Aware Tiered Memory for Agents. arXiv:2602.17913
A-MAC: Adaptive Memory Admission Control for LLM Agents. arXiv:2603.04549
MRAgent: Memory Is Reconstructed, Not Retrieved. ICLR virtual 10021254
Belief Engine: Bayesian Memory for Configurable Opinion Dynamics. ICLR virtual 10021252
MINJA: Memory Injection Attacks on LLM Agents via Query-Only Interaction. arXiv:2503.03704
SABER: Small Actions, Big Errors. arXiv:2512.07850
ShiftBench: Recovery of Agent Memory Under Distribution Shift. OpenReview CCSztIjmOy
DialSim: A Real-Time Simulator for Long-Term Dialogue. arXiv:2406.13144
SIRA: SuperIntelligent Retrieval Agent. arXiv:2605.06647
ERL: Experiential Reflective Learning. arXiv:2603.24639
WebCoach: Cross-Session Memory Guidance for Web Agents. arXiv:2511.12997
Distilling Feedback into Memory-as-a-Tool. arXiv:2601.05960
PROCED-MEM: Benchmarking Procedural Memory Retrieval. ICLR virtual 10021288

Don’t Ask the LLM to Track Freshness. arXiv:2606.01435
SAGE: a deterministic novelty gate for memory writes. arXiv:2605.30711
TOKI: a bitemporal operator algebra for contradiction resolution. arXiv:2606.06240
Kumiho: AGM belief-revision semantics for versioned graph memory. arXiv:2603.17244
Hindsight: epistemically typed memory with evolving confidence. arXiv:2512.12818
Memory-R1: RL-trained memory management, ACL 2026. arXiv:2508.19828
MemoryOS: OS-style tiered memory with heat-based eviction, EMNLP 2025. arXiv:2506.06326
Forgetful but Faithful: principled forgetting with regret bounds. arXiv:2512.12856
Memory Transfer Learning: insights transfer, traces do not. arXiv:2604.14004
An empirical study of agent-native memory systems (12 systems, 11 datasets). arXiv:2606.24775
The Coin Flip Judge: flip rates in LLM-as-judge evaluation. arXiv:2606.13685
Anatomy of Agentic Memory: a critique of memory evaluation. arXiv:2602.19320

LongMemEval. arXiv:2410.10813
MemoryAgentBench (conflict resolution and selective forgetting as first-class competencies). arXiv:2507.05257
BEAM (10M-token horizons). arXiv:2510.27246
STALE (do agents notice memories are no longer valid). arXiv:2605.06527
SWE-ContextBench (cross-session codebase tasks). arXiv:2602.08316
RealMem (project-oriented, evolving goals). arXiv:2601.06966

Tulving, Episodic and Semantic Memory, 1972. pdf
Anderson and Schooler, Reflections of the Environment in Memory, Psychological Science, 1991. doi:10.1111/j.1467-9280.1991.tb00174.x
Alchourrón, Gärdenfors, and Makinson, On the Logic of Theory Change, Journal of Symbolic Logic, 1985. doi:10.2307/2274239
Snodgrass, Developing Time-Oriented Database Applications in SQL, 1999. pdf
Lamport, Time, Clocks, and the Ordering of Events in a Distributed System, CACM, 1978. doi:10.1145/359545.359563
Kulkarni, Demirbas, Madappa, Avva, and Leone, Logical Physical Clocks and Consistent Snapshots, OPODIS 2014. pdf
Shapiro, Preguiça, Baquero, and Zawirski, Conflict-free Replicated Data Types, SSS 2011. hal:inria-00609399
Kleppmann, Wiggins, van Hardenberg, and McGranaghan, Local-First Software: You Own Your Data, in Spite of the Cloud, Onward! 2019. ink & switch
Kreps, The Log: What Every Software Engineer Should Know About Real-Time Data’s Unifying Abstraction, 2013. linkedin engineering
Cormack, Clarke, and Buettcher, Reciprocal Rank Fusion Outperforms Condorcet and Individual Rank Learning Methods, SIGIR 2009. doi:10.1145/1571941.1572114