From 0abfa8e96afffd75cdb6a985603dceb55a284427 Mon Sep 17 00:00:00 2001 From: YurenHao0426 Date: Wed, 8 Apr 2026 14:51:58 -0500 Subject: Appendix L: upgrade DFA+pen nudging/drift/trajectory to 3-seed (matches SB/CB) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Ran DFA+pen via cifar_resmlp.py on s123, s456. 3-seed: - acc 0.3601 ± 0.0014 (matches existing 0.363 from round 11) - deep cos +0.1518 ± 0.0110 (matches existing +0.155 ± 0.025) - deep nudge -5e-5 (was single-seed -6e-5) - train Δloss 0.095 ± 0.007 (was single-seed 0.104) - w2 drift 18.6 ± 0.5 (matches single-seed 18.8) - embed drift 94.6 ± 1.4 (matches single-seed 92.7) Single-seed s42 was within noise of the 3-seed mean across all 4 functional metrics, so the cos-vs-acc dissociation story is unchanged — Appendix L now just reports 3-seed for all 3 methods consistently. Co-Authored-By: Claude Opus 4.6 (1M context) --- paper/main.pdf | Bin 486620 -> 486667 bytes 1 file changed, 0 insertions(+), 0 deletions(-) (limited to 'paper/main.pdf') diff --git a/paper/main.pdf b/paper/main.pdf index 7d5dfea..da3099a 100644 Binary files a/paper/main.pdf and b/paper/main.pdf differ -- cgit v1.2.3