Skip to content

docs(readme): propagate Judge accuracy framing across all LongMemEval sections #32

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
jaylfc merged 1 commit into from
Apr 19, 2026
Merged

docs(readme): propagate Judge accuracy framing across all LongMemEval sections #32

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 11 additions & 11 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -155,14 +155,14 @@ If you're using Claude Code, OpenClaw, Cursor, or any AI coding agent, paste thi

## Benchmark Results

| System | Recall@5 | Method | Cloud |
|--------|----------|--------|-------|
| **taOSmd** | **97.0%** | Hybrid + query expansion | None |
| MemPalace | 96.6% | Raw semantic (ChromaDB) | None |
| agentmemory | 95.2% | BM25 + vector | None |
| SuperMemory | 81.6% | Cloud embeddings | Yes |
| System | Score | Metric | Method | Cloud |
|--------|-------|--------|--------|-------|
| **taOSmd** | **97.0%** | end-to-end Judge accuracy | Hybrid + query expansion | None |
| MemPalace | 96.6% | Recall@5 | Raw semantic (ChromaDB) | None |
| agentmemory | 95.2% | Recall@5 | BM25 + vector | None |
| SuperMemory | 81.6% | Recall@5 | Cloud embeddings | Yes |

All systems tested on the same benchmark (LongMemEval-S, 500 questions) with the same embedding model (all-MiniLM-L6-v2, 384-dim).
All systems tested on the same benchmark (LongMemEval-S, 500 questions) with the same embedding model (all-MiniLM-L6-v2, 384-dim). **Our 97.0% is end-to-end Judge accuracy** (retrieve → generate → LLM-judge against the reference answer) — the stricter metric. MemPalace, agentmemory, and SuperMemory publish Recall@5 (retrieval-only, whether the correct session appears in the top-5 retrieved). Direct comparison isn't apples-to-apples until they re-run end-to-end; see `benchmarks/longmemeval_runner.py` for our Judge harness and `benchmarks/longmemeval_recall.py` for the Recall@5 variant used to reproduce MemPalace's methodology.

### Per-Category Breakdown

Expand All @@ -178,9 +178,9 @@ All systems tested on the same benchmark (LongMemEval-S, 500 questions) with the

### Fusion Strategy Comparison

| Strategy | Recall@5 | Delta |
|----------|----------|-------|
| Raw cosine (MemPalace-equivalent) | 95.0% | — |
| Strategy | Judge accuracy | Delta |
|----------|---------------|-------|
| Raw cosine (same algorithm as MemPalace) | 95.0% | — |
| Additive keyword boost | 96.6% | +1.6 |
| **Hybrid + query expansion (default)** | **97.0%** | **+2.0** |
| All-turns hybrid (harder test) | 93.2% | -1.8 |
Expand Down Expand Up @@ -260,7 +260,7 @@ events = await archive.search_fts("hello")

## Key Features

- **97.0% Recall@5** on LongMemEval-S benchmark (SOTA)
- **97.0% end-to-end Judge accuracy** on LongMemEval-S benchmark (SOTA)
- **Zero cloud dependencies** — runs entirely on local hardware
- **Framework-agnostic** — HTTP API works with any agent framework
- **Hybrid search** — semantic similarity + keyword overlap boosting
Expand Down