✓✓ Beats tuned baseline 2026

Profile-Preserving Multislice Noise

Implementation & benchmark of arXiv:2607.13208 — The Action of the Lie Algebra $\mathfrak{sl}_n$ on Colored Graphs and Multicolored Johnson Graphs

Usefulness7/10

Difficulty3/10

Novelty7/10

Sampling Accuracy Sample efficiency Stability Graph nn Diffusion sampling Data augmentation Combinatorics Representation theory Probability Spectral theory

Source paper: The Action of the Lie Algebra $\mathfrak{sl}_n$ on Colored Graphs and Multicolored Johnson Graphs arXiv:2607.13208 ⓘ · analyzed Aug 30, 2026

AI-generated research hypothesis, automatically tested. Not peer-reviewed.

Idea description

For an input with exactly $\alpha_a$ occurrences of each state $a\in\{0,\ldots,n-1\}$, corrupt it by repeatedly swapping two positions with different states instead of independently resampling tokens. This defines a Markov process on the connected fixed-profile multislice, preserving global composition exactly and avoiding the distribution shift caused by ordinary categorical masking.

Formulas

$$(\mathbb{C}^{n})^{\otimes m},\qquad \alpha=(\alpha_0,\dots,\alpha_{n-1}),\qquad \sum_{a=0}^{n-1}\alpha_a=m.$$

$$A_\alpha=\sum_{1\le p<q\le m}\sum_{a\neq b}E_{ab}^{(p)}E_{ba}^{(q)}.$$

$$d_\alpha=\frac{1}{2}\left(m^2-\sum_{a=0}^{n-1}\alpha_a^2\right)=\sum_{a<b}\alpha_a\alpha_b.$$

$$P_\alpha=(1-\rho)I+\frac{\rho}{d_\alpha}A_\alpha.$$

Mathematical statement

The paper models a colored configuration with $m$ positions as a basis vector in $(\mathbb{C}^{n})^{\otimes m}$, where each position has state $a\in\{0,\ldots,n-1\}$ and profile $\alpha=(\alpha_0,\ldots,\alpha_{n-1})$ satisfies $\sum_a\alpha_a=m$. The multislice adjacency operator is the sum of exchanges of unequal states between two positions. Let $E_{ab}$ be the $n\times n$ matrix unit satisfying $E_{ab}e_b=e_a$ and $E_{ab}e_c=0$ for $c\neq b$, and let $E_{ab}^{(p)}$ act on position $p$. The exchange adjacency is $A_\alpha=\sum_{1\le p<q\le m}\sum_{a\neq b}E_{ab}^{(p)}E_{ba}^{(q)}$. Its degree is $d_\alpha=\frac12(m^2-\sum_a\alpha_a^2)=\sum_{a<b}\alpha_a\alpha_b$, because exactly that many unordered position pairs have unequal states. The random-transposition kernel is $P_\alpha=(1-\rho)I+(\rho/d_\alpha)A_\alpha$, with noise strength $\rho\in[0,1]$; every transition preserves $\alpha$.

Implementation notes

Integrate this at the data-loader or diffusion-noise layer for graph inputs represented by $m$ edge tokens, each taking one of $n$ states. First compute the profile $\alpha$ of each training example and store positions grouped by state. For one corruption step, choose an unordered pair of positions $(p,q)$ uniformly from all pairs with different states, then swap their state labels; repeat $k$ times, where $k$ is sampled from a schedule such as Uniform${\{0,\ldots,K\}}$ or chosen so that the expected number of swaps is $\rho d_\alpha$. This is an exact sampler for the random-transposition kernel $P_\alpha$; no matrix $A_\alpha$ should be materialized. Pseudocode is: x_noisy=x; for t in range(k): a=random_state_weighted_by(alpha); b=random_state_weighted_by(alpha excluding a); p=random position in bucket[a]; q=random position in bucket[b]; swap(x_noisy[p],x_noisy[q]); update the two buckets. Train a denoising autoencoder, graph classifier, or discrete diffusion model to predict the original configuration from x_noisy, conditioning on $\alpha$ and $k$. The mathematical quantities are the preserved profile, $d_\alpha$, and the exchange transition; the noise schedule and model parameterization are empirical choices. The first cheap experiment should use a small edge-state graph classifier on synthetic fixed-edge-count graph data, comparing independent categorical masking, standard random edge flips, and multislice swaps at equal corruption levels and equal model FLOPs. Measure validation accuracy, denoising cross-entropy, and performance when test profiles differ. Success is lower denoising loss and better fixed-profile generalization without profile violations; a direct invariant check must confirm that every corrupted sample has exactly the original count of every state.

Verification

Beats tuned baseline

Stage 1 · Toy mechanism gate: Passed ✓

Stage 2 · Mechanism transferred to benchmark: Confirmed ✓

Stage 2 · Practical benchmark result: Significant win ✓

Methodology: 8 paired seeds, permutation test (p<0.05) against a learning-rate-tuned baseline. How verification works

Stage 1 — Mechanism check agent confidence 9/10

Built an exact multislice swap sampler and a CUDA toy denoising experiment. The three mechanism predictions were confirmed: degree matched d=Σ_{a<b}α_aα_b exactly across four profiles, all 7-swap trials had zero profile violations, and empirical one-step self-transition probabilities matched 1−ρ within 0.004 across 12 profile/noise settings. In the toy denoiser at ρ=0.5, multislice noise achieved loss 0.673 and token accuracy 0.906 versus independent masking loss 0.911 and accuracy 0.653, so the promised profile-preservation effect is real in this setup.

Agent confidence: 9/10

Baseline: Independent categorical masking: cross-entropy 0.9106, token accuracy 0.6528
Idea: Profile-preserving multislice swaps: cross-entropy 0.6734, token accuracy 0.9060; degree errors 0, profile violations 0, kernel self-transition absolute error <=0.0034

Limitations: The ML comparison is a small synthetic fixed-profile sequence denoiser rather than a graph model or diffusion system. It tests one corruption strength and one profile, does not measure FLOPs or wall-clock speed, and does not evaluate generalization to shifted profiles or long-run multislice mixing.

How to run: python3 experiment.py

Stage 2 — Benchmark vs. tuned baseline (arXiv:2607.13208) Sep 2, 2026

Beats the tuned baseline by 51.2% (p=0.0081, wins 8 / 8 paired seeds; metric: lower is better).

Benchmark: Sequence modeling
Model: transformer_tiny
Paired seeds: 8
Baseline mean: 0.746
Idea mean: 0.364
Effect (Δ): -0.3821 (−51.2%; negative = idea better)
Wins: 8 / 8 paired seeds
p-value: 0.0081 (permutation test, 20 000 shuffles)
Smallest detectable effect: ±4.9%
Mechanism: Confirmed ✓
Practical effect: Helps
Baseline tuning: swept over 3 configs

Limitations:

The registered sequence track contains continuous forecast windows rather than categorical edge tokens, so quantile bins represent states. Only rho=0.5 was tested; the run used 1000 training samples, 300 test samples, and 8 epochs. No profile-shift, long-run mixing, wall-clock, FLOP, vision, or dynamics comparison was performed.

How to run: python3 stage2_multislice_bench.py

Verdict computed by deterministic test code from paired-seed statistics — not by the language model.

Stage 2 — Benchmark Aug 31, 2026 latest Worked ✓

Agent confidence: 8/10

Implemented Profile-Preserving Multislice Noise as quantile-state swaps over sequence windows, using the registered built-in sequence track and identical transformer systems for baseline and idea. Across 8 paired seeds, the idea reduced test MSE from 0.7460 to 0.3640, with paired delta -0.3821 and permutation p=0.0081; all 8 seeds improved. The fixed-bin mechanism signature was confirmed on benchmark batches.

Baseline: Independent per-token resampling, rho=0.5, best lr=0.001; 8-seed mean test MSE 0.746035
Idea: Quantile-state multislice swaps, rho=0.5, best lr=0.001; 8-seed mean test MSE 0.363980
Paired delta: -0.382055 test MSE; 8/8 idea wins; permutation p=0.0081

Benchmark result