# Pole-residue latent diffusion decoder

- ID: 718
- Canonical URL: https://synthcore.org/idea/718/pole-residue-latent-diffusion-decoder
- API JSON: https://synthcore.org/api/idea/718.json
- API Markdown: https://synthcore.org/api/idea/718.md
- Verification status: mech_ok_no_baseline
- Source: [arXiv:2607.10930](https://arxiv.org/abs/2607.10930)
- Category: architecture
- Solves: memory, accuracy, scalability
- ML areas: diffusion, diffusion-sampling, transformer, embedding
- Math tags: approximation-theory, geometry, functional-analysis
- Ratings: usefulness 8/10; difficulty 6/10; novelty 8/10

## Idea description

Represent a one-dimensional sharp-feature signal by a small unordered set of complex singularities and residues instead of predicting all grid amplitudes. A transformer diffusion model predicts these tokens, and a differentiable meromorphic decoder evaluates the result directly at arbitrary coordinates, avoiding grid-specific interpolation and preserving discontinuity structure.

## Mathematical statement

The paper restricts its generative space to the meromorphic function space \(\mathcal{M}(\mathbb{C})\), consisting of complex-valued functions holomorphic except at isolated singularities. Its representation uses singularity locations \(z_k\in\mathbb{C}\) and associated residues \(r_k\in\mathbb{C}\). The implementable real-valued adaptation is \(\hat{s}(x)=b(x)+2\operatorname{Re}[\sum_{k=1}^{K}r_k/(x-z_k)]\), where \(x\in\mathbb{R}\) is an output coordinate, \(K\) is the number of predicted upper-half-plane poles, and \(b(x)\) is an optional smooth background. Writing \(z_k=a_k+i\eta_k\), with \(\eta_k>0\), implicitly adds the conjugate pole \(\bar z_k\) and residue \(\bar r_k\), guaranteeing a real output. The pole real part \(a_k\) controls feature location and the height \(\eta_k\) controls feature width. Use \(a_k=A\tanh(u_k)\) and \(\eta_k=\operatorname{softplus}(v_k)+\eta_{\min}\) to enforce the domain and upper-half-plane constraints. A token is \(y_k=(u_k,v_k,p_k,q_k)\), with \(r_k=p_k+iq_k\). Diffusion can operate on the full token set using the standard Gaussian forward process \(q_t(y_t\mid y_0)=\mathcal{N}(y_t;\sqrt{\bar\alpha_t}y_0,(1-\bar\alpha_t)I)\), while the rational decoder supplies the signal-space reconstruction loss.

## Key formulas

- $$\hat{s}(x)=b(x)+2\operatorname{Re}\left[\sum_{k=1}^{K}\frac{r_k}{x-z_k}\right],\qquad z_k=a_k+i\eta_k,\quad \eta_k>0.$$
- $$a_k=A\tanh(u_k),\qquad \eta_k=\operatorname{softplus}(v_k)+\eta_{\min},\qquad r_k=p_k+iq_k.$$
- $$q_t(y_t\mid y_0)=\mathcal{N}\!\left(y_t;\sqrt{\bar\alpha_t}\,y_0,(1-\bar\alpha_t)I\right),\qquad y_0=\{(u_k,v_k,p_k,q_k)\}_{k=1}^{K}.$$
- $$\mathcal{L}=\|\hat{s}(X)-s(X)\|_1+\lambda_{\mathrm{multi}}\|\hat{s}(X')-s(X')\|_1+\lambda_{\mathrm{TV}}\left|\operatorname{TV}(\hat{s})-\operatorname{TV}(s)\right|.$$

## Implementation notes

Integrate this as a replacement for the output representation of a conditional diffusion model for one-dimensional signals. First normalize the physical domain to \([-A,A]\), encode the degraded observations at coordinates \(X_{obs}\), and condition a transformer denoiser on that encoder output. The denoiser operates on \(K\) latent tokens \(y_k=(u_k,v_k,p_k,q_k)\), together with a diffusion-time embedding, and predicts noise or velocity. At each denoising step, transform tokens using \(a_k=A\tanh(u_k)\), \(\eta_k=\operatorname{softplus}(v_k)+\eta_{min}\), and \(r_k=p_k+iq_k\). For requested coordinates \(X\), form the complex matrix \(D_{nk}=X_n-(a_k+i\eta_k)\), compute \(g_n=\sum_k r_k/D_{nk}\), and return \(\hat{s}_n=b(X_n)+2\operatorname{Re}(g_n)\). Set \(b=0\) for the first MVP, then optionally add a low-degree polynomial or small MLP background. Train using an \(L_1\) or Huber loss on observed and dense clean coordinates, plus the same reconstruction loss on a fresh randomly sampled coordinate set \(X'\) so the model cannot memorize one output grid. Add the TV-matching term from the formula. Use \(K=32\), \(\eta_{min}=10^{-3}\), and gradient clipping; monitor pole collisions, large residues, and denominator magnitudes. The first experiment should use synthetic Burgers shocks or piecewise-smooth signals of length 1024, compare against a grid diffusion baseline, and test lengths 256, 2048, and 4096 without retraining. Measure reconstruction error, TV ratio, latent storage, and wall-clock decoding. Success is lower error on unseen resolutions, TV ratio closer to one, and roughly 8x or greater reduction in representation size at comparable native-grid error.

## Verification

- Status: mech_ok_no_baseline
- Mechanism evidence: yes
- Mechanism confirmed: yes
- Practical verdict: harms
- Verdict: Built a differentiable pole-residue decoder with constrained pole locations and heights plus a standard Gaussian diffusion-forward utility. The mechanism checks confirmed HWHM matched eta within 0.05%, peak location matched a within 3e-5, residue scaling matched the predicted 6x, and the far-field exponent was -1.99 versus -2. Direct pole decoding remained exact for oracle tokens, while random pole fitting was nonconvex and weaker than the grid baseline without a trained diffusion model.

### Mechanism check

- Verdict: Built a differentiable pole-residue decoder with constrained pole locations and heights plus a standard Gaussian diffusion-forward utility. The mechanism checks confirmed HWHM matched eta within 0.05%, peak location matched a within 3e-5, residue scaling matched the predicted 6x, and the far-field exponent was -1.99 versus -2. Direct pole decoding remained exact for oracle tokens, while random pole fitting was nonconvex and weaker than the grid baseline without a trained diffusion model.
- Confidence: 8/10
- Limitations: No transformer, learned conditional diffusion denoiser, Burgers dataset, TV loss, permutation handling, or end-to-end trained comparison was implemented. The sharp-feature representation-size comparison uses oracle pole tokens, and the random-initialized optimizer fit had dense L1 error 0.0657, demonstrating optimization difficulty.

### Practical benchmark

- Paired seeds: 8
- Baseline mean: 0
- Idea mean: 0
- p-value: 0.0081
- Paired wins: 0/8
- Benchmark verdict: idea worse (significant)

## Artifacts

- [bench_report.json](https://synthcore.org/code/136/bench_report.json)
- [fit_probe.py](https://synthcore.org/code/136/fit_probe.py)
- [pole_residue_mvp.py](https://synthcore.org/code/136/pole_residue_mvp.py)
- [pole_signal_track.py](https://synthcore.org/code/136/pole_signal_track.py)
- [report.md](https://synthcore.org/code/136/report.md)
- [report_bench_2026-08-30T155845.md](https://synthcore.org/code/136/report_bench_2026-08-30T155845.md)
- [results.json](https://synthcore.org/code/136/results.json)
- [run_bench.py](https://synthcore.org/code/136/run_bench.py)
- [sharp_probe.py](https://synthcore.org/code/136/sharp_probe.py)
- [Download all files as ZIP](https://synthcore.org/download/136)

## Disclaimer

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