summaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
2026-04-08OPTION F polish pass: update abstract, §1 ¶2, §6 ¶3, §7 ¶2 with round 3...YurenHao0426
2026-04-08Round 40 §4 update: cos-vs-acc 3-part proposition (Observation / Inference /...YurenHao0426
2026-04-08Round 38 CB+penalty multi-seed: confirm clean cos-vs-acc dissociation across ...YurenHao0426
2026-04-08Appendix G + J: 3-seed depth scan L=4 validation + SB/CB random_targets 100ep...YurenHao0426
2026-04-08Appendix J: add EP random_targets full 100ep convergence ||h_L||=2085, acc=0.081YurenHao0426
2026-04-08Round 38 paper update: §4 + §5 + new Appendix K with SB+penalty 3-seed resultYurenHao0426
2026-04-08Appendix H: H2 ablation now 3-seed (s42, s123, s456) — multi-seed validates...YurenHao0426
2026-04-08Round 38: add --penalty_lam flag to cifar_resmlp.py for Mode 2 cross-method testYurenHao0426
2026-04-08Fix bibtex citation key: refinetti2023align -> refinetti2023aligning (matches...YurenHao0426
2026-04-08Fix §3 vanilla DFA comparison: use 100ep audit value 0.306±0.006 (matches t...YurenHao0426
2026-04-08Fix precision: ||g_L|| = 7.2e-4 (mean), not 7.4e-4 (rounding)YurenHao0426
2026-04-08Round 37 OPTION C: §3 compression — preserve causal structure, drop per-ro...YurenHao0426
2026-04-08Round 36: upgrade (b) wording + add EP random-target neg control to §3YurenHao0426
2026-04-08Add 100ep trajectory confirmations to Appendix I (H2) and Appendix J (random ...YurenHao0426
2026-04-08Round 35: SB and CB also show data-agnostic Mode 1 growth on random targetsYurenHao0426
2026-04-08Round 34 random-target ablation: Mode 1 fires under random labels tooYurenHao0426
2026-04-08Round 32+33 H2 ablation: add no_residual_add flag; falsify residual-as-cause ...YurenHao0426
2026-04-08Add depth-sweep evidence to §5 + Appendix H from existing d=512 L=2,4,6,8,12...YurenHao0426
2026-04-08Round 31: fill §6 Recommended Protocol + §7 Discussion prose; v2 content-co...YurenHao0426
2026-04-08Round 30: fill in §5 Intervention and Cross-Arch prose (4 paragraphs) via codexYurenHao0426
2026-04-08Round 29: fill in §4 Failure Mode 2 prose (3 paragraphs) via codexYurenHao0426
2026-04-08Round 28: fill in §3 Failure Mode 1 prose (4 paragraphs) via codexYurenHao0426
2026-04-08Round 27: fill in §2 Audit prose (4 paragraphs) via codexYurenHao0426
2026-04-08Round 26: fill in §1 Introduction prose (3 paragraphs) via codexYurenHao0426
2026-04-08Fill in tables 1-3 + generate figures 2/4/5 from existing dataYurenHao0426
2026-04-08v2 skeleton from round 25: section structure now matches round 23YurenHao0426
2026-04-08Archive failed v1 draft as v1_rejected.tex; remove main.tex/main.pdfYurenHao0426
2026-04-08Compile paper PDF + fix bibstyle for tectonicYurenHao0426
2026-04-08Paper main.tex: add §5.4 λ sweep dissociation tableYurenHao0426
2026-04-08λ sweep on penalty strength: lam ∈ {1e-4, 1e-2, 1e-1} cos + rho resultsYurenHao0426
2026-04-08First draft of NeurIPS 2026 E&D paperYurenHao0426
2026-04-08Add perturbation correlation metric calibrationYurenHao0426
2026-04-08PAPER_OUTLINE: add 6th validation (perturbation correlation triangulation)YurenHao0426
2026-04-08Extend perturbation audit to vanilla early-epoch checkpointsYurenHao0426
2026-04-08EVIDENCE_SUMMARY: add 6th validation (perturbation correlation triangulation)YurenHao0426
2026-04-08Add perturbation correlation audit (round 19's recommended alt metric)YurenHao0426
2026-04-08EVIDENCE_SUMMARY: §4 fully rewritten under locked two-distinct-modes framingYurenHao0426
2026-04-08PAPER_OUTLINE: round 20 language tightening + 5 validation summaryYurenHao0426
2026-04-08Multi-seed vanilla DFA early-epoch cos: lock-in for round 19 disambiguationYurenHao0426
2026-04-08BP+penalty control result: mode 2 (intrinsic credit quality) confirmed REALYurenHao0426
2026-04-08Add BP+penalty control (round 19's #4 critical experiment)YurenHao0426
2026-04-08PAPER_OUTLINE: §4 rewrite under 'two distinct failure modes' framingYurenHao0426
2026-04-083-seed multi-seed verification of penalized DFA deep cos = +0.17YurenHao0426
2026-04-08DISAMBIGUATION: vanilla DFA early-epoch checkpoints + cos measurementYurenHao0426
2026-04-08Add vanilla DFA early-epoch checkpoint training (round 19 disambiguation)YurenHao0426
2026-04-08Add null calibration script: training-Bs vs fresh-Bs cos on penalized DFAYurenHao0426
2026-04-08MAJOR: penalized DFA deep-layer cosine is +0.17, NOT zeroYurenHao0426
2026-04-08Add PAPER_OUTLINE.md: §1-§6 draft reflecting round 17 + 18YurenHao0426
2026-04-08Add penalty lambda 3-seed summary script + checkpoint save in penalty testYurenHao0426
2026-04-08Add penalty λ sweep figure: shows λ-dependence of (d) verdictYurenHao0426
generated by cgit v1.2.3 (git 2.25.1) at 2026-06-15 05:57:53 +0000