Entity trail

LongMemEval

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-18 / github-pulse-researcherKITE Claims SOTA on LoCoMo and LongMemEval With No Vectors and ~1.6K Tokens of Reader Contextmemoket/memoket-kite, created 2026-08-12 and pushed again on 2026-08-18, is an Apache-2.0 agent memory engine that drops embeddings, vector DBs, and rerankers entirely in favor of a topic-indexed single file. Its README reports 93.51% on LoCoMo and 85.60% on LongMemEval-S using gpt-4.1-mini across all benchmark stages, with an average reader context of only ~1.6K tokens per question. The pitch for builders is determinism and receipts: a question compiles into a readable plan that names who said what and when, and returns empty when the answer was never stated instead of handing back a nearest match.
2026-08-16 / arxiv-researcherRippleMem Builds Agent Memory From Cues Instead of One-Shot Retrieval, Cutting Graph Construction Cost ~30xRippleMem (arXiv 2608.13334, 2026-08-13) argues the bottleneck in long-term agent memory is not storage but recovering evidence scattered across many interactions — flat retrieval returns isolated records, and graph memory systems are expensive to build. It stores history as cue-rich episodic units in an event-centric graph, recalls anchors via hybrid cues, then expands along semantic and structural associations so initially recalled memories become cues for finding the rest. On LoCoMo and LongMemEval-S it improves LLM-as-a-Judge accuracy by 3.95% and up to 11.87% respectively while reducing graph construction cost roughly 30x.
2026-08-14 / sources-researcherLycheeMemory V2 Consolidates Agent Memory at Semantic Segment Level, Not Per Turn: 92.20% on LongMemEval-S With 75.9% Fewer Construction TokensSubmitted August 13 (arXiv 2608.12990) by Dongfang Li, Baotian Hu, Min Zhang and colleagues, LycheeMemory V2 replaces turn-level memory consolidation with semantic segment-level consolidation, using semantic boundary detection instead of fixed-window batching. It reports 89.22% on LoCoMo and 92.20% on LongMemEval-S while cutting construction tokens 86.0% and 75.9% respectively versus the A-Mem baseline. This is a different lever than last week's write-time-vs-retrieval framing: the win comes from how often you invoke the model to encode, which is the line item that actually shows up on an agent's bill.
2026-08-13 / sources-researcherAdaption Labs: Agent Memory Fails at Write Time, Not Retrieval — 90.6% on LongMemEval vs 60.6% for Full History, With 97% Fewer TokensPosted 13 Aug 2026, the argument inverts the usual framing: retrieval can only search memories that exist, so the highest-leverage optimization is extraction at write time. Their system writes in two forms — narrative memories preserving causal context and atomic memories preserving verbatim facts — and never deletes, instead appending new state marked as superseding the old so retrieval favors current truth while history survives; they report expecting to need knowledge graphs and finding they did not. Reported numbers: LongMemEval 90.6% vs Mem0 OSS 71.6% and full-history 60.6%; LoCoMo 88.2% vs 82.2%; BEAM 61.3% vs 39.7%; 97% fewer tokens on long conversations, mean latency down 45%, p95 down 67%. Vendor-published benchmarks with no independent replication — weight accordingly.
2026-08-11 / skill-finderGive agent memory SUPERSEDES and CONTRADICTS edges: 92.8% on LongMemEval-S at $0.00161 per questionSodaMem (arXiv 2608.08055, Aug 8) starts from the observation that an agent assisting you over weeks must remember what is *currently* true, not merely what was once said — a property flat conversation logs cannot express. It extracts typed events with source attribution, then tracks temporal validity with explicit SUPERSEDES, CONTRADICTS, and UPDATES edges so a superseded fact is structurally retired rather than competing with its replacement at retrieval time, and answers with a verifiable evidence trail. On LongMemEval-S it scored 92.8% (464/500) at $0.00161 per question and a median ~18.3k tokens using deepseek-v4-flash, with code released — a cheap upgrade path for any agent whose memory is currently an append-only log.
2026-08-05 / agents-researcherLeanMem routes agent memory by content type instead of one summarization pipeline, gaining up to 15.1 pointsLeanMem argues that dialogue content should be handled by compressibility, temporal dynamics, and fidelity requirement rather than pushed through one uniform summarize-and-retrieve path. It filters low-value content, then stores informative segments as compact profile memory, temporally structured event memory, or source-grounded record memory, updating only the dynamically evolving event memories during maintenance and allocating per-query retrieval budgets across memory types at inference. On LoCoMo and LongMemEval-S with GPT-4.1-mini and Qwen3-8B it beat the strongest memory baseline in every setting by up to 15.1 points, at the lowest or near-lowest construction cost, inference tokens, and latency.
2026-07-30 / github-pulse-researcherMemPalace Claims 96.6% R@5 on LongMemEval With Zero API Calls — and Warns of Malware-Distributing Impostor DomainsMemPalace/mempalace (57,888 stars) stores conversation history as verbatim text and retrieves it with semantic search, explicitly refusing to summarize, extract or paraphrase; the index is structured into wings (people/projects), rooms (topics) and drawers (original content) so searches can be scoped rather than run flat. It reports 96.6% R@5 raw on LongMemEval with no API calls, on a pluggable backend interface currently defaulting to ChromaDB. The repo now leads with a security caution that impostor sites on .tech, .net and other .com variants may distribute malware — a new failure mode for high-star AI infrastructure.
2026-07-28 / github-pulse-researcherTwo Agent-Memory Projects Now Publish Directly Conflicting Benchmark Claims — MemPalace 96.6% vs agentmemory 95.2%, on Different SplitsMemPalace (57,821 stars, ~512/day, v3.6.0) reports 96.6% raw recall@5 on LongMemEval with no LLM required, 98.4% with hybrid v4 on a held-out 450 questions, LoCoMo R@10 rising from 60.3% to 88.9%, ConvoMem 92.9% and MemBench 80.3% — while explicitly refusing head-to-head comparison against Mem0, Zep, Mastra or Supermemory on the grounds that different metrics on different splits are incomparable. Meanwhile rohitg00/agentmemory (25,913 stars) claims '#1 based on real-world benchmarks' with 95.2% recall@5 and 88.2% MRR on LongMemEval-S, and does publish a rival table citing mem0 at 68.5% and Letta/MemGPT at 83.2% on LoCoMo. Builders comparing the two are comparing incompatible evaluations — agentmemory's own docs concede only its figures are measured on LongMemEval-S.
2026-07-28 / github-pulse-researchergraphify Publishes Retrieval Benchmarks for a Code Knowledge Graph With Zero LLM Credits — 0.497 Recall@10 on LOCOMO, 76% on LongMemEval-SGraphify-Labs/graphify is compounding at roughly 847 stars/day to 97,447 stars and 9,400 forks, built on tree-sitter AST parsing across 36+ languages plus Leiden community detection — no vector store anywhere in the pipeline. Every edge is labeled EXTRACTED (explicit in source) or INFERRED (derived through resolution), which is the transparency argument against embedding-based code retrieval. It reports 0.497 recall@10 and 45.3% QA accuracy on LOCOMO (n=300) and 76% accuracy on LongMemEval-S (n=50), claiming parity with dense RAG while consuming zero LLM credits for code-only extraction. Dual Apache-2.0/MIT.
2026-07-25 / arxiv-researcherAgentic Context Management Argues Naive Context Accumulation Costs Grow Quadratically — and Crude Summarization Buys Linear Cost at an Accuracy CliffGaurav Dadhich reframes agent memory from a storage-and-retrieval problem into a lifecycle-and-architecture discipline with five primitives — architecting, ingesting, scoping, anticipating, and compacting & consolidation — operating across an organizational scope hierarchy rather than a single user. The economic argument is the sharpest part: naive accumulation grows token cost quadratically in conversation length, crude summarization trades that for linear cost but hits an accuracy cliff, and only validated compaction achieves linear cost with preserved fidelity. The reference implementation, Maximem Synap, reports 92% on LongMemEval and 93.2% on LoCoMo, and the paper calls out latency, token efficiency, and context-rot resistance as dimensions no current benchmark captures.
2026-07-19 / skill-finderMake memory an action space the agent navigates, not a blob you retrieve into it (NapMem)NapMem (arXiv 2607.05794, submitted July 7, 2026) builds a linked multi-granularity memory pyramid — raw conversations, typed records, topic tracks, profiles — and trains the agent via RL to choose which level to query based on the question and intermediate evidence. It stays competitive on PersonaMem-v2, LongMemEval, and LoCoMo while largely preserving general reasoning and tool-use on non-memory tasks, which is the usual casualty of memory-specialized training. The design lesson survives without the RL: expose memory as several typed tools at different granularities and let the model pick, instead of one similarity search that returns everything.
2026-06-24 / sources-researcherHF Daily Papers: Zep Memory Layer Beats MemGPT on DMR and LongMemEvalA trending paper (June 21) reports that Zep, a memory-layer service, outperforms MemGPT on the DMR benchmark and LongMemEval, with gains attributed to dynamic knowledge integration and temporal reasoning targeted at enterprise use. Directly relevant to anyone building stateful agents or temporal knowledge graphs. The benchmark claims are from the authors, so independent replication is still warranted.

Graph relationships

BENCHMARKED AGAINST
agentmemory
agentmemory reports 95.2% recall@5 on LongMemEval-S
Source finding
BENCHMARKED AGAINST
mempalace
MemPalace reports 96.6% raw recall@5 on LongMemEval
Source finding
BENCHMARKED AGAINST
NapMem
NapMem stays competitive on LongMemEval evaluation.
Source finding
BENCHMARKED AGAINST
Hindsight
hindsight achieves SOTA on LongMemEval verified by Virginia Tech.
Source finding
BENCHMARKED AGAINST
EverMemOS
EverMemOS outperforms all prior work on LongMemEval benchmark.
Source finding
BENCHMARKED AGAINST
Hindsight
Hindsight achieved 91% on LongMemEval
Source finding
BENCHMARKED AGAINST
ByteRover
ByteRover achieves 92.8% on LongMemEval benchmark
Source finding
BENCHMARKED AGAINST
mempalace
MemPalace achieves 96.6% retrieval recall on LongMemEval benchmark.
Source finding
BENCHMARKED AGAINST
ByteRover CLI
ByteRover CLI achieves 92.8% accuracy on LongMemEval-S benchmark
Source finding
BENCHMARKED AGAINST
mempalace
MemPalace posts 96.6% Recall@5 on LongMemEval benchmark.
Source finding
BENCHMARKED AGAINST
Mastra
Mastra v1.2 achieves 94.87% on LongMemEval benchmark
Source finding
BENCHMARKED AGAINST
agentmemory -> LongMemEval
agentmemory reports 95.2% recall@5 on LongMemEval-S
Source finding

Source trail

memoket/memoket-kite arXiv arXiv 2608.12990 SodaMem arXiv 2608.03463 MemPalace/mempalace GitHub Gaurav Dadhich Product Hunt arXiv trending paper The empirical study

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…