The Continuity Layer
Primer — Evidence Library

Evidence Library

The public artifacts this project’s continuity claims and category map are built on. Each entry: what it is, what it supports, and what’s still open. The org-level Kenotic Labs story lives in Field Watch Entry 001 — these cards cover only the artifacts themselves. The broader field beyond this project’s own sources lives in Signals from the Field — and for the points where those sources actively conflict, where the field currently disagrees.

Start Here routeStep 4 of 4
From Lab to evidence

Lab defines what has to be tested. Evidence Library shows the receipts, references, and source material behind the claims. This supports proof-before-claim discipline; it does not mean the proof is already complete.

Evidence artifacts

ATANT v1.0 — evaluation framework paper

Continuity evaluation framework

One of the first public attempts to define and score continuity properties directly, rather than treating memory recall as a proxy for continuity.

Claim supported

That continuity properties (persistence, update handling, disambiguation, and similar) can be operationalized into a testable framework.

Caveats

Single-author framework, single narrative corpus, self-evaluated to date on Kenotic Labs' own reference implementation (NURA) — independent replication is an open question. Published by Kenotic Labs, whose commercial work this project also profiles — read as the field's originating argument, not independent corroboration.

arXiv:2604.06710

ATANT v1.1 — positioning paper

Continuity evaluation framework

Positions continuity evaluation explicitly against memory, long-context, and agentic-memory benchmarks, arguing continuity measures a different thing than any of them alone.

Claim supported

That existing memory/long-context benchmarks don't cover what continuity evaluation is trying to measure.

Caveats

Single-author framework, single narrative corpus, self-evaluated to date on Kenotic Labs' own reference implementation (NURA) — independent replication is an open question. Published by Kenotic Labs, whose commercial work this project also profiles — read as the field's originating argument, not independent corroboration.

arXiv:2604.10981

The Continuity Layer (thesis paper)

Continuity thesis / positioning

Makes the broader argument for why intelligence needs an architecture for what it carries forward, independent of any one product.

Claim supported

The category-level argument that carried-forward state is an architectural requirement, not an optional feature.

Caveats

A positioning paper, not an empirical evaluation — read alongside the ATANT papers, which carry the testable claims. Published by Kenotic Labs, whose commercial work this project also profiles — read as the field's originating argument, not independent corroboration.

arXiv:2604.17273

ATANT — GitHub repository

Open benchmark code

The evaluation code is public and inspectable, which is rarer than it should be in this field.

Claim supported

Reproducibility in principle — the code is available to run.

Caveats

Public code doesn't by itself mean independent parties have run it yet; that remains open.

github.com/Kenotic-Labs/ATANT

ATANT v1.0 corpus — HuggingFace dataset

Benchmark dataset

The narrative corpus underlying the ATANT evaluations is public and downloadable.

Claim supported

Transparency of the underlying test material.

Caveats

Single corpus — breadth across different kinds of narratives and domains is untested.

huggingface.co/datasets/Kenotic-Labs/ATANTV1.0-corpus

“RAG Doesn’t Solve Hallucination. What Actually Does.”

Field commentary

Argues the specific limitation of RAG this project also maps under “Memory Is Not Continuity” — that retrieval-of-similar isn't verification-of-current.

Claim supported

The RAG-versus-hallucination distinction as a live, publicly argued position in the field, not just this project's framing.

Caveats

Vendor-published essay — read as argument, not independent research.

kenoticlabs.com/insights/rag-hallucination

“Why Do 80% of AI Agent Projects Fail?”

Field commentary

A public claim about agent-project failure rates tied to state loss between steps and sessions — relevant to the “Why This Matters Now” framing.

Claim supported

That agent failure is commonly attributed to lost state, not model capability, in public industry discussion.

Caveats

The 80% figure is the vendor's framing; this project has not independently verified the underlying methodology.

kenoticlabs.com/insights/ai-agent-failure

“Why Does Character AI Forget Everything?”

Field commentary

A concrete, named-product example of the memory/continuity gap, useful as a plain-language illustration.

Claim supported

That consumer AI products with memory features still lose continuity across sessions in practice.

Caveats

Single-product example; not a systematic survey of consumer AI memory failures.

kenoticlabs.com/insights/character-ai-memory

LOCOMO

Continuity/memory benchmark

One of the earliest widely-cited benchmarks for very long-term conversational memory, testing multi-session and temporal QA.

Claim supported

That multi-session memory recall is measurable and that current systems have quantifiable gaps on it.

Caveats

Measures conversational recall, not the fuller continuity property set (disambiguation, reconstruction, model independence) this project tracks.

arXiv:2402.17753; github.com/snap-research/locomo

LongMemEval

Continuity/memory benchmark

500-question benchmark covering information extraction, multi-session reasoning, temporal reasoning, knowledge updates, and abstention — the closest existing public benchmark to several of this project's seven continuity properties.

Claim supported

That update-handling and abstention are measurable and that current chat assistants have specific, quantified weaknesses there.

Caveats

Benchmark-specific findings; results are per-model and change as models update.

arXiv:2410.10813 (ICLR 2025); github.com/xiaowu0162/longmemeval

BEAM

Long-context/memory benchmark

Generates coherent synthetic conversations up to 10M tokens with probing questions, designed to force genuine retrieval-based memory rather than reward long-context stuffing.

Claim supported

That long-context alone is distinguishable from actual memory/retrieval performance at scale.

Caveats

Synthetic conversations — real-world conversational structure may differ.

github.com/mohammadtavakoli78/BEAM (ICLR 2026); huggingface.co/datasets/Mohammadta/BEAM

MemoryBench

Memory/continual-learning benchmark

Evaluates memory and continual learning in LLM systems via simulated accumulated user feedback across multiple domains and languages.

Claim supported

That continual, feedback-driven memory update is measurable across domains, not just within one narrow task.

Caveats

A same-named but distinct project exists at github.com/supermemoryai/memorybench — this entry refers specifically to the THUIR/ICML benchmark; do not conflate the two when citing.

arXiv:2510.17281; github.com/THUIR/MemoryBench (ICML 2026 Spotlight)

RULER

Long-context benchmark

A synthetic, configurable suite measuring the real usable context size of long-context models, not just their advertised context window.

Claim supported

That advertised long-context length and effective usable context are frequently different things — directly relevant to the “long context isn't continuity” contrast.

Caveats

Measures long-context retrieval capability, not continuity properties like update handling or model independence.

arXiv:2404.06654; github.com/NVIDIA/RULER

Zep evals

Vendor-published memory system evaluation

A public, technical account of a temporal knowledge-graph approach to agent memory, evaluated on DMR and LongMemEval.

Claim supported

That graph-structured, time-aware memory can improve measured performance on existing memory benchmarks.

Caveats

Vendor-authored; a headline result reported in the paper was publicly disputed and later corrected by a third party (github.com/getzep/zep-papers/issues/5). Treat performance claims as self-reported pending independent reproduction.

arXiv:2501.13956 (“Zep: A Temporal Knowledge Graph Architecture for Agent Memory”)

Letta / MemGPT evals

Vendor-published memory system evaluation

A public leaderboard scoring models on agentic memory read/write/update tasks, from the team behind the MemGPT/Letta agent memory framework.

Claim supported

That agentic memory read/write/update behavior varies significantly by underlying model and is measurable.

Caveats

Company-run benchmark, blog-published rather than peer-reviewed; methodology has not been independently replicated.

letta.com/blog/letta-leaderboard; letta.com/blog/benchmarking-ai-agent-memory
Continue the proof trail

Once the source trail is visible, continue through evaluation standards, declared tests, local practice, or the underlying model.