Section F of the Feature Backlog: implicit diffusion: calibration and posterior samplers; likelihoods, polynomial and algebraic factors. Open rows only; a row that is done moves to Backlog Done 2026-11 with its id. Status, priority, ids and the column needs are explained in Feature Backlog, which also holds the dependency graph.
Sources: original to this repository; the sources of each row are in its source column.
| id | feature | status | needs | priority | source | why |
|---|---|---|---|---|---|---|
| F1 | Graded block levels in ImplicitDiffusion: a fixed noise level per block that rises with the block’s age (old blocks nearly clean, recent ones noisier, free ones annealed), generalising today’s annealed-or-clean blocks | proposed | — | low | LenticulumLM’s Incremental Next-Token Prediction (an idea, 2026-11-05) | the schedule of Rolling Diffusion and Diffusion Forcing as a query option; holding clamps clean already helps (implicit.md §9); whether untrained-for schedules help is open |
| F2 | Calibration benchmarks for implicit relations | partial: circle, lemniscate, unbalanced rings (2026-11-05, lib/VariationalDiffusion.jl/examples/calibration.md); a pitchfork and a redundant arm remain | — | medium | both reviews (LP-023) | baseline measured: every off-domain query returns a root, the lemniscate’s recall is at best 0.52, the 10% ring is never found |
| F3 | Calibration remedies: off-relation detection (the full score’s norm, an energy threshold, a round trip), annealing the field levels, ridge residuals | partial: round_trip and ridge_residual (VariationalDiffusion answers.jl), and the round trip, the ridge residual, coverage and annealing measured on the benchmark (2026-11-12 and 13, calibration.md “Remedies”, LP-030 Findings of the review of round_trip, LP-033 Findings of the review of the ridge residual); an energy threshold remains | F2 | medium | both reviews; calibration.md | measured: both checks remove every off-domain root; coverage removes 9 of 11 and 96 of 99 spurious roots on the circle and the two circles with no loss of recall, but costs recall at the lemniscate’s crossing (0.52 to 0.31); the round trip also rejects true roots where the relation is parallel to the clamped coordinate (circle recall 0.86 to 0.74), which a ridge residual would not; annealing by continuation = (16, 4) lowered recall everywhere and left the lemniscate with no root; the ridge residual removes every off-domain root and loses no matched root at the low levels (spurious 11 to 3, 36 to 16, 99 to 27) |
| F4 | Diffusion posterior samplers, as query methods of ImplicitDiffusion beside root finding, RED-Diff (reddiff.jl), ProxDM (proxdm.jl) and replacement/RePaint (implicit_sample). Each is checked against the exact posterior of GaussianMixtureEps under a linear-Gaussian measurement, and compared on the calibration benchmark (rows below) | proposed | F2 | medium | Daniel, 2026-11-05 (“whatever useful sampling method appears”) | today the library samples a query by replacement or optimises it (RED-Diff, ProxDM); a noisy or nonlinear measurement needs guided or particle samplers |
| F4a | — DPS: the reverse SDE guided by through Tweedie’s estimate (arXiv:2209.14687; chung2023dps in refs.bib) | proposed | F2 | medium | as above | the standard baseline; needs the network’s VJP, which the adjoints provide |
| F4b | — ΠGDM: pseudoinverse guidance for linear measurements (Song, Vahdat, Mardani & Kautz, ICLR 2023; not on arXiv, metadata to check on OpenReview) | proposed | F2 | low | as above | cheaper and better-conditioned guidance for linear operators |
| F4c | — DiffPIR as a sampler: half-quadratic splitting with the noise predictor’s Tweedie denoiser as the prior’s proximal step, then re-noising (arXiv:2305.08995; zhu2023diffpir) | proposed | F2 | low | as above | prox_infer already does the splitting with a proximal network; this uses an ordinary noise predictor |
| F4d | — DDRM: SVD-based sampling for linear inverse problems (arXiv:2201.11793) | proposed | F2 | low | as above | exact handling of a linear operator’s singular directions |
| F4e | — DAPS: decoupled noise annealing (arXiv:2407.01521) | proposed | F2 | low | as above | large moves early, against DPS’s local corrections |
| F4f | — Twisted diffusion sampler: sequential Monte Carlo with twisting functions, asymptotically exact conditional sampling (arXiv:2306.17775) | proposed | F2 | medium | as above | the one exact sampler of the family as particles grow: an oracle for the others, and weighted samples are a SampleBelief the graph can use |
| F4g | — MCG: manifold-constrained gradient (arXiv:2206.00941) | proposed | F2 | low | as above | a correction to DPS-style guidance |
| F5 | Heteroscedastic and heavy-tailed likelihoods | proposed | — | low | ChatGPT review §5 | |
| F6 | Polynomial factor: higher-degree ideal quotients, missing-data EM | planned | — | low | polyfit.md | |
| F7 | Algebraic factors: MLE existence by LP, Markov bases, splitting large tables | planned | — | low | L-SPEC-004 Algebraic Factors Toric and Grassmannian open, §2.4 | |
| F8 | Polarity-dependent energies (allowed by the interface, used by no factor) | proposed | — | low | L-SPEC-001 The Factor Interface open |