---
kind: concept
slug: feynman-kac-correctors
title: Feynman-Kac correctors
summary: >-
  Sampling a target built from several pretrained diffusion models, or tilted toward a reward, by carrying log-weights along transported particles instead of retraining. Covers the general Feynman-Kac potential, the generalised version for arbitrary terminal functions of N agents with reward steering, and its known weakness: weight degeneracy. Author-reported results only, from one paper.
tags: [sampling, diffusion, guidance, model-composition, importance-weighting]
papers: ["2610.40383"]
related: [diffusion-models, importance-weighting, classifier-free-guidance, classifier-guidance, reward-proportional-sampling, flow-matching]
status: draft
updated_at: 2026-11-21
---

## Overview

Suppose you want samples from a density that is a function of several
pretrained diffusion models, for example a ratio, a sum or a mixture of their
marginals, or such a density tilted toward a reward. Naively adding the models'
scores gives biased samples, because the intermediate distributions along the
generative path are then wrong [arxiv:2610.40383 §1].

The shared idea is to keep a set of particles, move each one with a velocity
field, and accumulate a log-weight per particle from a potential along its
path. The weighted particles then estimate the target through self-normalised
importance weights. Resampling is optional [arxiv:2610.40383 §2.2]. The
invariant all variants share is that the final normalised weights estimate the
target, so no model is retrained [arxiv:2610.40383 §3].

The paper builds the target along a geometric path from an initial function h0
to the target h1 of the agents' final marginals, and the potential is what
makes the path's distribution exact [arxiv:2610.40383 §3].

## Variants

### Earlier Feynman-Kac correctors (as described by the paper)

The paper's introduction and background describe an earlier family of
Feynman-Kac correctors that treats specific target families exactly: annealed
distributions, products of experts, and classifier-free-guidance-style
geometric averages [arxiv:2610.40383 §1] [arxiv:2610.40383 §2.2]. These
methods are not in this wiki with a card; the paper's account is the only
source on this page for them. The paper says its general result recovers that
earlier method as a special case [arxiv:2610.40383 §1].

### Generalised Feynman-Kac correctors (GFKC)

Card: [GFKC](/papers/2610.40383), arXiv:2610.40383.

- **Target.** A density proportional to h1 evaluated on the N agents'
  marginals at t = 1, where h1 is positive, differentiable and integrable.
  Products, geometric averages, annealed distributions, mixtures and ratios
  are the paper's examples [arxiv:2610.40383 §3].
- **Potential.** The paper gives the potential in closed form for an arbitrary
  unifying velocity built from the agents' velocities. Its sampler then uses
  the velocity of agent i for the particles assigned to agent i
  [arxiv:2610.40383 §3].
- **Log-densities along the path.** Because a particle moves with one agent's
  velocity but each agent's density must still be evaluated, the usual
  log-density change-of-variables equation is generalised. It adds a
  divergence term and an inner product of velocity differences with the scores
  [arxiv:2610.40383 §3].
- **Reward steering.** For a reward r and steering parameter kappa > 0 the
  target becomes h1 times exp(r / kappa). The change is one extra term in the
  potential. The reward needs a terminal sample, so each particle is projected
  to t = 1 with its agent's sampler at chosen steering times [arxiv:2610.40383
  §3] [arxiv:2610.40383 §4.1].
- **Reported experiments (author-reported).** Two Gaussian agents with exact
  scores, where the authors show close agreement with the analytic target and
  mass shifts under steering [arxiv:2610.40383 fig. 1] [arxiv:2610.40383 fig. 3].
  Two MNIST digit-class agents trained as small diffusion models, where the
  sum and ratio targets pick out the expected digits and steering moves samples
  toward a rotated 9 [arxiv:2610.40383 fig. 6] [arxiv:2610.40383 fig. 10].

### Reward-guided inference for a single model

The paper contrasts its method with reward-guided inference-time diffusion,
which steers one model at a time [arxiv:2610.40383 §2.3]. That method is named
here and linked on arXiv only: [arXiv:2501.09685](https://arxiv.org/abs/2501.09685).
Nothing from its content is used on this page.

## When to use which

| Situation | Use | Why |
|---|---|---|
| Target is an arbitrary positive function of several pretrained diffusion models (ratio, sum, mixture) and retraining is not an option | GFKC weighted particles | The general framework covers these targets without retraining [arxiv:2610.40383 §3] |
| Target is annealed, a product of experts, or a classifier-free-guidance-style geometric average | Earlier Feynman-Kac corrector, or GFKC as its special case | The earlier method is exact for these families; the paper says GFKC recovers it [arxiv:2610.40383 §1] |
| Want samples tilted toward a reward on the output of a model you already have | GFKC steering with kappa | Adds one term to the potential; the paper reports steering effects for two experiments [arxiv:2610.40383 §3] [arxiv:2610.40383 fig. 4] |
| Need a reliable effective sample size, or a high-dimensional target with few particles | Treat with caution; add resampling and check weights | Log-weights can differ by thousands, so almost all mass sits on one agent [arxiv:2610.40383 fig. 7]; the paper names this as the main open problem [arxiv:2610.40383 §5] |
| Only the naive score sum is available and an exact target is not needed | Naive combination, knowing it is biased | The paper argues the sum does not reproduce the intermediate distributions [arxiv:2610.40383 §1] |

Constraints that the cited results support:

- **Model access.** The method needs each agent's score and velocity, and
  divergence estimates along the path [arxiv:2610.40383 §3].
- **Compute.** Each step evaluates every agent's score and a divergence per
  agent. The MNIST experiments use a Hutchinson estimator with ten probes
  [arxiv:2610.40383 §4.2]. The paper reports no runtime comparison in the text
  this page is drawn from.
- **Numerics.** All computation is in the log domain. The MNIST experiments
  run in double precision and clip log-densities from above at every step
  [arxiv:2610.40383 §4.2]. The clip value is not stated in the paper.
- **Resampling.** Algorithm 1 includes a resampling step at a set of times,
  but the paper does not state that set for its experiments [arxiv:2610.40383
  alg. 1] [arxiv:2610.40383 §4].

## Implementation sketch

The common core of the Feynman-Kac corrector family, with the GFKC-specific
parts marked where they differ.

```pseudocode
# Generic Feynman-Kac corrector  [arxiv:2610.40383 §2.2]
# Particles x_m with log-weights logw_m; velocity v and potential g
# are the two model-dependent parts.

# Start from the common prior N(0, I), uniform log-weights  [arxiv:2610.40383 §3]
x = sample_prior(M)
logw = zeros(M)

for t in time_grid:
    # Potential at the current state, accumulated over the step  [§2.2]
    logw = logw + dt * g(x, t)
    # Transport each particle with the velocity  [§2.2]
    x = x + dt * v(x, t)
    # Optional: resample when the weights degenerate  [arxiv:2610.40383 alg. 1]
    if resample_now(t):
        x, logw = resample(x, logw)
    # Subtracting a shared constant from logw leaves normalised weights unchanged

# Self-normalised estimate of the target  [arxiv:2610.40383 §2.2]
w = softmax(logw)
```

Where GFKC differs:

- The velocity v is chosen per particle from the agent it is assigned to,
  and the potential g is the closed-form expression for the chosen unifying
  velocity [arxiv:2610.40383 §3].
- Each agent's log-density is tracked along the path with the generalised
  equation above, and it is clipped from above at every step [arxiv:2610.40383 §3]
  [arxiv:2610.40383 §4.2].
- For steering, the potential gains a reward term evaluated on the terminal
  sample of each particle, projected to t = 1 at the chosen steering times
  [arxiv:2610.40383 §4.1].

## Contradictions

Only one card lists this concept, so no two papers disagree. Within that
paper, the released code differs from the text on several points that change
the sampler: the noise term, the index used for steering times, and the
resampling set. The card lists each one with its file reference, and this page
does not resolve them: see [the paper card](/papers/2610.40383), section
"Paper vs code".

## Papers

- [GFKC](/papers/2610.40383): generalises the Feynman-Kac corrector to arbitrary terminal functions of N diffusion agents, adds reward steering, and reports Gaussian and MNIST experiments.
