Test-Time Policy Adaptation for Enhanced Multi-Turn Interactions with LLMs

• URL: http://arxiv.org/abs/2509.23166v1
• Date: Sat, 27 Sep 2025 07:46:15 GMT
• Title: Test-Time Policy Adaptation for Enhanced Multi-Turn Interactions with LLMs
• Authors: Chenxing Wei, Hong Wang, Ying He, Fei Yu, Yao Shu,
• Abstract summary: We introduce Test-Time Policy Adaptation for Multi-Turn Interactions (T2PAM)<n>We first propose a new paradigm: T2PAM, which utilizes user feedback as a reward signal to estimate a latent optimal policy aligned with user preferences.<n>We then introduce Optimum-Referenced One-Step Adaptation (ROSA), a lightweight algorithm that operationalizes T2PAM.
• Score: 20.892283201423048
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) employ multi-turn interaction as a fundamental paradigm for completing complex tasks. However, their performance often degrades in extended interactions, as they are typically trained on static, single-turn data, which hinders their ability to adapt to real-time user feedback. To address this limitation, we first propose a new paradigm: Test-Time Policy Adaptation for Multi-Turn Interactions (T2PAM), which utilizes user feedback from the ongoing interaction as a reward signal to estimate a latent optimal policy aligned with user preferences, then updates a small subset of parameters to steer the model toward this policy, ultimately enabling efficient in-conversation self-correction. We then introduce Optimum-Referenced One-Step Adaptation (ROSA), a lightweight algorithm that operationalizes T2PAM. ROSA guides the model parameters toward a theoretical optimal policy in a single, efficient update step, avoiding costly iterative gradient-based optimization and minimizing computational overhead. We provide a rigorous theoretical analysis guaranteeing that the policy of ROSA converges to the preference of user as the number of interactions increases. Extensive experiments on challenging benchmark demonstrate that ROSA achieves significant improvements in both task effectiveness and efficiency.

Related papers

• Words & Weights: Streamlining Multi-Turn Interactions via Co-Adaptation [55.938648534942665]
  Test-time policy adaptation for multi-turn interactions (T2PAM) is essential for aligning Large Language Models (LLMs) with dynamic user needs during inference time.<n>We propose ROSA2, a framework that reformulates interaction as a joint optimization problem over the heterogeneous space of Words and Weights.
  arXiv  Detail & Related papers  (2026-03-02T02:16:20Z)
• Q-learning with Adjoint Matching [58.78551025170267]
  We propose Q-learning with Adjoint Matching (QAM), a novel TD-based reinforcement learning (RL) algorithm.<n>QAM sidesteps two challenges by leveraging adjoint matching, a recently proposed technique in generative modeling.<n>It consistently outperforms prior approaches on hard, sparse reward tasks in both offline and offline-to-online RL.
  arXiv  Detail & Related papers  (2026-01-20T18:45:34Z)
• MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization [56.074760766965085]
  Group-Relative Policy Optimization has emerged as an efficient paradigm for aligning Large Language Models (LLMs)<n>We propose MAESTRO, which treats reward scalarization as a dynamic latent policy, leveraging the model's terminal hidden states as a semantic bottleneck.<n>We formulate this as a contextual bandit problem within a bi-level optimization framework, where a lightweight Conductor network co-evolves with the policy by utilizing group-relative advantages as a meta-reward signal.
  arXiv  Detail & Related papers  (2026-01-12T05:02:48Z)
• GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization [133.27496265096445]
  We show how to apply Group Relative Policy Optimization under multi-reward setting without examining its suitability.<n>We then introduce Group reward-Decoupled Normalization Policy Optimization (GDPO), a new policy optimization method to resolve these issues.<n>GDPO consistently outperforms GRPO, demonstrating its effectiveness and generalizability for multi-reward reinforcement learning optimization.
  arXiv  Detail & Related papers  (2026-01-08T18:59:24Z)
• ACPO: Adaptive Curriculum Policy Optimization for Aligning Vision-Language Models in Complex Reasoning [17.928214942495412]
  ACPO employs a dynamic curriculum that orchestrates a principled transition from a stable, near on-policy exploration phase to an efficient, off-policy exploitation phase.<n>We conduct extensive experiments on a suite of challenging multimodal reasoning benchmarks, including MathVista, LogicVista, and MMMU-Pro.<n>Results demonstrate that ACPO consistently outperforms strong baselines such as DAPO and PAPO, achieving state-of-the-art performance, accelerated convergence, and superior training stability.
  arXiv  Detail & Related papers  (2025-10-01T09:11:27Z)
• Value-Free Policy Optimization via Reward Partitioning [0.08192907805418585]
  We introduce Reward Partitioning Optimization (RPO), a new method for single-trajectory reinforcement learning.<n>RPO normalizes observed rewards using a approach estimated directly from data.<n>We validate RPO on scalar-feedback language modeling tasks using Flan-T5 encoder-decoder models.
  arXiv  Detail & Related papers  (2025-06-16T17:06:27Z)
• Multi-Objective Preference Optimization: Improving Human Alignment of Generative Models [15.799929216215672]
  We introduce the Multi-Objective Preference Optimization (MOPO) algorithm, which frames alignment as a constrained KL-regularized optimization.<n>Unlike prior work, MOPO operates directly on pairwise preference data, requires no point-wise reward assumption, and avoids prompt-context engineering.
  arXiv  Detail & Related papers  (2025-05-16T05:58:26Z)
• C-MORL: Multi-Objective Reinforcement Learning through Efficient Discovery of Pareto Front [9.04360155372014]
  Constrained MORL is a seamless bridge between constrained policy optimization and MORL.<n>Our algorithm achieves more consistent and superior performances in terms of hypervolume, expected utility, and sparsity on both discrete and continuous control tasks.
  arXiv  Detail & Related papers  (2024-10-03T06:13:56Z)
• Bridging and Modeling Correlations in Pairwise Data for Direct Preference Optimization [75.1240295759264]
  We propose an effective framework for Bridging and Modeling Correlations in pairwise data, named BMC.<n>We increase the consistency and informativeness of the pairwise preference signals through targeted modifications.<n>We identify that DPO alone is insufficient to model these correlations and capture nuanced variations.
  arXiv  Detail & Related papers  (2024-08-14T11:29:47Z)
• Interactive Test-Time Adaptation with Reliable Spatial-Temporal Voxels for Multi-Modal Segmentation [56.70910056845503]
  Multi-modal test-time adaptation (MM-TTA) adapts models to an unlabeled target domain by leveraging the complementary multi-modal inputs in an online manner.<n>Previous MM-TTA methods for 3D segmentation suffer from two major limitations: 1) unstable frame-wise predictions caused by temporal inconsistency, and 2) consistently incorrect predictions that violate the assumption of reliable modality guidance.<n>This work introduces a comprehensive two-fold framework: Latte++ that better suppresses the unstable frame-wise predictions with more informative geometric correspondences, and Interactive Test-Time Adaptation (ITTA), a flexible add-on to empower effortless human feedback
  arXiv  Detail & Related papers  (2024-03-11T06:56:08Z)
• Towards Efficient Exact Optimization of Language Model Alignment [93.39181634597877]
  Direct preference optimization (DPO) was proposed to directly optimize the policy from preference data. We show that DPO derived based on the optimal solution of problem leads to a compromised mean-seeking approximation of the optimal solution in practice. We propose efficient exact optimization (EXO) of the alignment objective.
  arXiv  Detail & Related papers  (2024-02-01T18:51:54Z)

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.