AvAtar: Learning to Align via Active Optimal Transport

• URL: http://arxiv.org/abs/2605.24395v1
• Date: Sat, 23 May 2026 04:40:21 GMT
• Title: AvAtar: Learning to Align via Active Optimal Transport
• Abstract summary: We propose a principled active alignment framework for optimal transport alignment called AvAtar.<n>We quantify the informativeness of a candidate by measuring its gradient-based impact on the global alignment result.<n>By encoding the global alignment result via effective utility functions, AvAtar is applicable to general alignment problems under the OT framework.
• Score: 70.846866626571
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Alignment plays a fundamental role in many machine learning problems, such as multi-network analysis, multimodal learning, and point cloud registration. Recent works increasingly leverage optimal transport (OT) for distributional alignment, whose effectiveness largely depends on sparse supervision that is hard or costly to obtain in practice. Existing works, however, largely overlook how to actively acquire high-quality supervision to improve their alignment performance under OT frameworks. In this paper, we propose a principled active alignment framework for optimal transport alignment called AvAtar. We quantify the informativeness of a candidate by measuring its gradient-based impact on the global alignment result, computed as the gradient propagation from the global alignment result to all possible supervisions of the candidate through the entropy-regularized OT formulation. While differentiating through OT is challenging given its constrained nature, we leverage the adjoint-state method to reformulate the computation to a linear system solvable by the conjugate gradient method with linear complexity and guaranteed convergence. By encoding the global alignment result via effective utility functions, AvAtar is applicable to general alignment problems under the OT framework. Extensive experiments on three representative alignment tasks demonstrate the effectiveness, scalability, and generalizability of the proposed AvAtar.

Related papers

• Event-Driven Reinforcement Learning Enables Long-Horizon Control in Semiconductor Fabrication [1.3810107538833278]
  We propose a deep reinforcement learning framework for multi-objective policy optimization at this scale.<n>Specifically, we formulate control as a central-agent problem, where a core policy coordinates system-wide decisions.<n>We develop a tailored event-driven temporal-difference formulation that remains general and can be integrated with various policy optimization methods.
  arXiv  Detail & Related papers  (2026-06-09T11:08:45Z)
• OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL [63.388513841293616]
  Existing forgery detection methods fail to handle the interleaved text, images, and videos prevalent in real-world misinformation.<n>To bridge this gap, this paper targets to develop a unified framework for omnibus vision-language forgery detection and grounding.<n>We propose textbf OmniVL-Guard, a balanced reinforcement learning framework for omnibus vision-language forgery detection and grounding.
  arXiv  Detail & Related papers  (2026-02-11T09:41:36Z)
• Iterative Refinement of Flow Policies in Probability Space for Online Reinforcement Learning [56.47948583452555]
  We introduce the Stepwise Flow Policy (SWFP) framework, founded on the key insight that discretizing the flow matching inference process via a fixed-step Euler scheme aligns it with the variational Jordan-Kinderlehrer-Otto principle from optimal transport.<n>SWFP decomposes the global flow into a sequence of small, incremental transformations between proximate distributions.<n>This decomposition yields an efficient algorithm that fine-tunes pre-trained flows via a cascade of small flow blocks, offering significant advantages.
  arXiv  Detail & Related papers  (2025-10-17T07:43:51Z)
• COPO: Consistency-Aware Policy Optimization [17.328515578426227]
  Reinforcement learning has significantly enhanced the reasoning capabilities of Large Language Models (LLMs) in complex problem-solving tasks.<n>Recently, the introduction of DeepSeek R1 has inspired a surge of interest in leveraging rule-based rewards as a low-cost alternative for computing advantage functions and guiding policy optimization.<n>We propose a consistency-aware policy optimization framework that introduces a structured global reward based on outcome consistency.
  arXiv  Detail & Related papers  (2025-08-06T07:05:18Z)
• Joint Optimal Transport and Embedding for Network Alignment [66.49765320358361]
  We propose a joint optimal transport and embedding framework for network alignment named JOENA.<n>With a unified objective, the mutual benefits of both methods can be achieved by an alternating optimization schema with guaranteed convergence.<n>Experiments on real-world networks validate the effectiveness and scalability of JOENA, achieving up to 16% improvement in MRR and 20x speedup.
  arXiv  Detail & Related papers  (2025-02-26T17:28:08Z)
• Learning Dynamic Representations via An Optimally-Weighted Maximum Mean Discrepancy Optimization Framework for Continual Learning [16.10753846850319]
  Continual learning allows models to persistently acquire and retain information.<n> catastrophic forgetting can severely impair model performance.<n>We introduce a novel framework termed Optimally-Weighted Mean Discrepancy (OWMMD), which imposes penalties on representation alterations.
  arXiv  Detail & Related papers  (2025-01-21T13:33:45Z)
• Learning Reward and Policy Jointly from Demonstration and Preference Improves Alignment [58.049113055986375]
  We develop a single stage approach named Alignment with Integrated Human Feedback (AIHF) to train reward models and the policy.<n>The proposed approach admits a suite of efficient algorithms, which can easily reduce to, and leverage, popular alignment algorithms.<n>We demonstrate the efficiency of the proposed solutions with extensive experiments involving alignment problems in LLMs and robotic control problems in MuJoCo.
  arXiv  Detail & Related papers  (2024-06-11T01:20:53Z)
• PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback [106.63518036538163]
  We present a novel unified bilevel optimization-based framework, textsfPARL, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning. Our framework addressed these concerns by explicitly parameterizing the distribution of the upper alignment objective (reward design) by the lower optimal variable. Our empirical results substantiate that the proposed textsfPARL can address the alignment concerns in RL by showing significant improvements.
  arXiv  Detail & Related papers  (2023-08-03T18:03:44Z)
• Compositionality of Linearly Solvable Optimal Control in Networked Multi-Agent Systems [27.544923751902807]
  We discuss the methodology of generalizing the optimal control law from learned component tasks to unlearned composite tasks on Multi-Agent Systems (MASs) The proposed approach achieves both the compositionality and optimality of control actions simultaneously within the cooperative MAS framework in both discrete- and continuous-time in a sample-efficient manner.
  arXiv  Detail & Related papers  (2020-09-28T20:21:48Z)

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.