4 Commits

Author SHA1 Message Date
claw d8a030d9e9 finding #43: opus + narrow framing for contradiction detection
Tests the open question from Finding #39: does Opus's internal reasoning
depth suffice for self-contradiction verification?

Key result: wrong question. Opus finds a different CLASS of contradiction
than GPT-5. GPT-5 finds specification conflicts (statement comparison).
Opus finds logical impossibilities (deductive rule interaction). Neither
dominates — they don't overlap. Sonnet remains unreliable (~33% precision).

Document tested: escalation-policy.md (228 lines)
Models: GPT-5, Claude Opus 4.6, Claude Sonnet 4.6
2026-05-07 16:05:14 -07:00
claw 0c632c255a finding #39: narrow framing does not close Sonnet-GPT-5 gap for semantic consistency
Tested open question from Finding #5: does narrow framing give Sonnet
GPT-5-level semantic analysis?

Result: NO. Narrow framing changes WHAT Sonnet looks for (redirects from
gaps to contradictions) but not HOW WELL it reasons. Sonnet narrow found
3 contradictions but only 1 was genuine (2 were analytical errors/misreads).
GPT-5 found 4 all-genuine findings with precise reasoning.

Key insight: framing controls scope, not reasoning depth. For tasks
requiring logical verification (contradictions, race conditions, invariant
violations), reasoning tokens are necessary — framing alone is insufficient.

Updated open-questions.md: marked Sonnet+narrow as answered, added new
question about Opus+narrow for contradiction detection.
2026-05-07 09:26:08 -07:00
claw d8ddbc9861 mark adversarial ensemble question as answered (finding #35) 2026-05-06 21:29:35 -07:00
Rodin 1b108ff66e Initial publish: 29 findings, 6 prompts, methodology, open questions
Full comparative analysis of GPT-5, Claude Opus 4.6, Claude Sonnet 4.6,
GPT-4.1, and GPT-4.1 Mini on analytical tasks (not coding).

Contents:
- findings/ALL-FINDINGS.md — complete 3,249-line research log with all
  29 findings, methodology notes, and open questions
- prompts/ — 6 exact prompts used across experiments
- methodology.md — experimental setup and evaluation criteria
- open-questions.md — unanswered questions for future work
- README.md — overview and summary table

Key findings:
- Cross-document consistency: Opus is 2.4x faster with more findings
- Gap-finding: GPT-5 reasoning tokens find domain-specific gaps
- Race conditions: Opus excels at temporal interaction reasoning
- Bias detection: Signal-to-noise ratio > model capability
- Adversarial analysis: GPT-5 exhaustive, Opus qualitatively different

Signed-off-by: Rodin
2026-05-05 19:13:03 -07:00