floq: Training Critics via Flow-Matching for Scaling Compute in Value-Based RL

• URL: http://arxiv.org/abs/2509.06863v2
• Date: Thu, 23 Oct 2025 14:41:11 GMT
• Title: floq: Training Critics via Flow-Matching for Scaling Compute in Value-Based RL
• Authors: Bhavya Agrawalla, Michal Nauman, Khush Agrawal, Aviral Kumar,
• Abstract summary: floq is an approach that parameterizes the Q-function using a velocity field and trains it using techniques from flow-matching.<n>Floq improves performance by nearly 1.8x across a suite of challenging offline RL benchmarks and online fine-tuning tasks.
• Score: 26.288205235851887
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A hallmark of modern large-scale machine learning techniques is the use of training objectives that provide dense supervision to intermediate computations, such as teacher forcing the next token in language models or denoising step-by-step in diffusion models. This enables models to learn complex functions in a generalizable manner. Motivated by this observation, we investigate the benefits of iterative computation for temporal difference (TD) methods in reinforcement learning (RL). Typically they represent value functions in a monolithic fashion, without iterative compute. We introduce floq (flow-matching Q-functions), an approach that parameterizes the Q-function using a velocity field and trains it using techniques from flow-matching, typically used in generative modeling. This velocity field underneath the flow is trained using a TD-learning objective, which bootstraps from values produced by a target velocity field, computed by running multiple steps of numerical integration. Crucially, floq allows for more fine-grained control and scaling of the Q-function capacity than monolithic architectures, by appropriately setting the number of integration steps. Across a suite of challenging offline RL benchmarks and online fine-tuning tasks, floq improves performance by nearly 1.8x. floq scales capacity far better than standard TD-learning architectures, highlighting the potential of iterative computation for value learning.

Related papers

• What Does Flow Matching Bring To TD Learning? [28.717975688380488]
  Flow matching can be effective for scalar Q-value function estimation in reinforcement learning (RL)<n>We show that their success is not explained by distributional RL, as explicitly modeling return distributions can reduce performance.<n>We argue that the use of integration for reading out values and dense velocity supervision at each step of this integration process for training improves TD learning via two mechanisms.
  arXiv  Detail & Related papers  (2026-03-04T17:51:30Z)
• When Bayesian Tensor Completion Meets Multioutput Gaussian Processes: Functional Universality and Rank Learning [53.17227599983122]
  Functional tensor decomposition can analyze multi-dimensional data with real-valued indices.<n>We propose a rank-revealing functional low-rank tensor completion (RR-F) method.<n>We establish the universal approximation property of the model for continuous multi-dimensional signals.
  arXiv  Detail & Related papers  (2025-12-25T03:15:52Z)
• DiRL: An Efficient Post-Training Framework for Diffusion Language Models [54.405206032785706]
  Diffusion Language Models (dLLMs) have emerged as promising alternatives to Auto-Regressive (AR) models.<n>Existing methods suffer from computational inefficiency and objective mismatches between training and inference.<n>We introduce DiRL, an efficient post-training framework that tightly integrates FlexAttention-accelerated blockwise training with LMDeploy-optimized inference.
  arXiv  Detail & Related papers  (2025-12-23T08:33:19Z)
• SoFlow: Solution Flow Models for One-Step Generative Modeling [10.054000663262618]
  Flow Models (SoFlow) is a framework for one-step generation from scratch.<n>Flow Matching loss allows our models to provide estimated velocity fields during training.<n>Our models achieve better FID-50K scores than MeanFlow models on the ImageNet 256x256 dataset.
  arXiv  Detail & Related papers  (2025-12-17T18:10:17Z)
• Learning Distinguishable Representations in Deep Q-Networks for Linear Transfer [0.9558392439655014]
  We propose a novel deep Q-learning approach that introduces a regularization term to reduce positive correlations between feature representation of states.<n>We demonstrate the efficacy of our approach in improving transfer learning performance and thereby reducing computational overhead.
  arXiv  Detail & Related papers  (2025-09-29T15:44:35Z)
• Efficient Machine Unlearning via Influence Approximation [75.31015485113993]
  Influence-based unlearning has emerged as a prominent approach to estimate the impact of individual training samples on model parameters without retraining.<n>This paper establishes a theoretical link between memorizing (incremental learning) and forgetting (unlearning)<n>We introduce the Influence Approximation Unlearning algorithm for efficient machine unlearning from the incremental perspective.
  arXiv  Detail & Related papers  (2025-07-31T05:34:27Z)
• Aligning Frozen LLMs by Reinforcement Learning: An Iterative Reweight-then-Optimize Approach [65.6966065843227]
  Iterative Reweight-then-IRO is a framework that performs RL-style alignment of a frozen base model without touching its parameters.<n>At test time, the value functions are used to guide the base model generation via a search-based optimization process.<n> Notably, users can apply IRO to align a model on their own dataset, similar to OpenAI's reinforcement fine-tuning (RFT)
  arXiv  Detail & Related papers  (2025-06-21T21:49:02Z)
• Diffusion-Model-Assisted Supervised Learning of Generative Models for Density Estimation [10.793646707711442]
  We present a framework for training generative models for density estimation. We use the score-based diffusion model to generate labeled data. Once the labeled data are generated, we can train a simple fully connected neural network to learn the generative model in the supervised manner.
  arXiv  Detail & Related papers  (2023-10-22T23:56:19Z)
• The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute [66.84421705029624]
  We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours. We pre-process an existing large, diverse, and high-quality dataset of books that surpasses existing academic benchmarks in quality, diversity, and document length. This work also provides two baseline models: a feed-forward model derived from the GPT-2 architecture and a recurrent model in the form of a novel LSTM with ten-fold throughput.
  arXiv  Detail & Related papers  (2023-09-20T10:31:17Z)
• Training Energy-Based Normalizing Flow with Score-Matching Objectives [36.0810550035231]
  We present a new flow-based modeling approach called energy-based normalizing flow (EBFlow) We demonstrate that by optimizing EBFlow with score-matching objectives, the computation of Jacobian determinants for linear transformations can be entirely bypassed.
  arXiv  Detail & Related papers  (2023-05-24T15:54:29Z)
• Benchmarking Learning Efficiency in Deep Reservoir Computing [23.753943709362794]
  We introduce a benchmark of increasingly difficult tasks together with a data efficiency metric to measure how quickly machine learning models learn from training data. We compare the learning speed of some established sequential supervised models, such as RNNs, LSTMs, or Transformers, with relatively less known alternative models based on reservoir computing.
  arXiv  Detail & Related papers  (2022-09-29T08:16:52Z)
• Deep Equilibrium Optical Flow Estimation [80.80992684796566]
  Recent state-of-the-art (SOTA) optical flow models use finite-step recurrent update operations to emulate traditional algorithms. These RNNs impose large computation and memory overheads, and are not directly trained to model such stable estimation. We propose deep equilibrium (DEQ) flow estimators, an approach that directly solves for the flow as the infinite-level fixed point of an implicit layer.
  arXiv  Detail & Related papers  (2022-04-18T17:53:44Z)
• Self Normalizing Flows [65.73510214694987]
  We propose a flexible framework for training normalizing flows by replacing expensive terms in the gradient by learned approximate inverses at each layer. This reduces the computational complexity of each layer's exact update from $mathcalO(D3)$ to $mathcalO(D2)$. We show experimentally that such models are remarkably stable and optimize to similar data likelihood values as their exact gradient counterparts.
  arXiv  Detail & Related papers  (2020-11-14T09:51:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.