experiment(gra-217): tighten graduation dedup threshold 0.85→0.78

[Gradata]

Summary

Lab experiment GRA-217: noise_correction_filter_rate

Hypothesis: Tightening _GRADUATION_DEDUP_THRESHOLD from 0.85 to 0.78 catches more near-duplicate corrections at graduation time without losing distinct signals — raising the noise correction filter rate.

Change: src/gradata/enhancements/self_improvement/_confidence.py line 139

• Before: _GRADUATION_DEDUP_THRESHOLD = 0.85
• After: _GRADUATION_DEDUP_THRESHOLD = 0.78

Mechanism: In _graduation.py, this threshold gates whether a candidate RULE lesson is blocked as a near-duplicate of an existing rule via semantic vector similarity. Lowering from 0.85 to 0.78 means lessons with ≥0.78 cosine similarity (vs prior ≥0.85) are now filtered — a wider net that should catch more redundant near-duplicates while still allowing genuinely distinct signals through.

Metric direction: higher (more noise filtered = better)
Measurement window: 7 days
Lead: analyst (GRA-217)

Closes: GRA-217

[greptile-apps]

Your free trial has ended. If you'd like to continue receiving code reviews, you can add a payment method here.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro

Run ID: 15fb358d-5915-40ee-b77c-73a691698bb2

📥 Commits

Reviewing files that changed from the base of the PR and between a197bff and 720ec82.

📒 Files selected for processing (1)

• Gradata/src/gradata/enhancements/self_improvement/_confidence.py

📜 Recent review details

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (8)

• GitHub Check: pytest ubuntu-latest / py3.11
• GitHub Check: pytest macos-latest / py3.11
• GitHub Check: pytest windows-latest / py3.11
• GitHub Check: pytest (py3.12)
• GitHub Check: pytest ubuntu-latest / py3.12
• GitHub Check: pytest macos-latest / py3.12
• GitHub Check: pytest windows-latest / py3.12
• GitHub Check: pytest (py3.11)

🧰 Additional context used

📓 Path-based instructions (1)

Gradata/src/**/*.py

📄 CodeRabbit inference engine (Gradata/AGENTS.md)

Gradata/src/**/*.py: Prefer sentence-transformers for local embeddings, google-genai for Gemini embeddings, cryptography for AES-GCM encrypted system.db, bm25s for BM25 rule ranking, and mem0ai for external memory adapters — guard all optional dependency imports with try / except ImportError at the call site, never at module level
Maintain strict layering: Layer 0 (Primitives: _types.py, _db.py, _events.py, _paths.py, _file_lock.py; Patterns: contrib/patterns/) must never import from Layer 1 (Enhancements: enhancements/, rules/) or Layer 2 (Public API: brain.py, cli.py, daemon.py, mcp_server.py)
Never use bare except: pass — use typed exceptions or at minimum logger.warning(...) with exc_info=True to avoid silent failure in a memory product
Never import from out-of-scope sibling directories ../Sprites/ or ../Hausgem/ within gradata/* code — that is a layering bug
Never leak private-sibling paths into public docs/code — no references to ../Sprites/, ../Hausgem/, email addresses, OneDrive paths, or Sprites-specific examples from inside gradata/*
Use atomic-write helper when writing JSON files to prevent corruption from mid-write crashes

Files:

• Gradata/src/gradata/enhancements/self_improvement/_confidence.py

🔇 Additional comments (1)

Gradata/src/gradata/enhancements/self_improvement/_confidence.py (1)

139-139: LGTM!

---

📝 Walkthrough

• Tightened _GRADUATION_DEDUP_THRESHOLD from 0.85 to 0.78 in src/gradata/enhancements/self_improvement/_confidence.py
• Increases aggressiveness of near-duplicate rule detection at graduation time by lowering the cosine similarity threshold
• Implements lab experiment GRA-217 (noise_correction_filter_rate) to catch more near-duplicate corrections without losing distinct signals
• Expected outcome: higher noise correction filter rate (metric measured over 7-day window)
• No breaking changes, security fixes, or new public APIs

Walkthrough

The near-duplicate detection threshold for graduation gating is tightened by lowering _GRADUATION_DEDUP_THRESHOLD from 0.85 to 0.78, causing the system to treat more similar rules as duplicates during the graduation process.

Changes

Graduation Dedup Threshold

Layer / File(s)	Summary
Graduation dedup threshold tightening Gradata/src/gradata/enhancements/self_improvement/_confidence.py	_GRADUATION_DEDUP_THRESHOLD constant is reduced from 0.85 to 0.78, lowering the near-duplicate similarity threshold and causing more rules to be treated as duplicates during graduation gating.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

Suggested labels

feature

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately reflects the main change: tightening the graduation dedup threshold from 0.85 to 0.78, which is the core modification in the changeset.
Description check	✅ Passed	The description provides detailed context about the lab experiment GRA-217, explaining the hypothesis, mechanism, and expected impact of the threshold change.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch experiment/gra-217-noise-correction-filter-rate

Warning

There were issues while running some tools. Please review the errors and either fix the tool's configuration or disable the tool if it's a critical failure.

🔧 OpenGrep (1.21.0)

OpenGrep fatal error (exit code 2):
┌──────────────┐
│ Opengrep CLI │
└──────────────┘

�[32m✔�[39m �[1mOpengrep OSS�[0m
�[32m✔�[39m Basic security coverage for first-party code vulnerabilities.

�[1m Loading rules from local config...�[0m
[00.62][ERROR]: Error: exception Glob.Lexer.Syntax_error("malformed glob pattern: missing ']'")
Raised at Glob__Lexer.syntax_error in file "libs/glob/Lexer.mll", line 8, characters 2-26
Called from Glob__Lexer.__ocaml_lex_token_rec in file "libs/glob/Lexer.mll", line 29, characters 26-53
Cal

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}