HiPP-Prune: Hierarchical Preference-Conditioned Structured Pruning for Vision-Language Models

• URL: http://arxiv.org/abs/2603.06270v1
• Date: Fri, 06 Mar 2026 13:31:54 GMT
• Title: HiPP-Prune: Hierarchical Preference-Conditioned Structured Pruning for Vision-Language Models
• Abstract summary: HiPP-Prune is a hierarchical preference-conditioned structured pruning framework that treats pruning as conditional resource allocation.<n>HiPP-Prune makes plan-level decisions by factorizing decisions into an overall sparsity budget and a layer-wise allocation.
• Score: 2.1418081649543157
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Pruning vision-language models (VLMs) for efficient deployment is challenging because compression can affect not only task utility but also visual grounding, often amplifying object hallucinations even at the same sparsity level. We present HiPP-Prune, a hierarchical preference-conditioned structured pruning framework that treats pruning as conditional resource allocation under multiple objectives. HiPP-Prune makes plan-level decisions: a single policy invocation outputs a global pruning blueprint by factorizing decisions into an overall sparsity budget and a layer-wise allocation, enabling queryable trade-offs via a user-specified preference vector. To account for VLM-specific failure modes, our policy state integrates a visual sensitivity signal derived from attention flow between vision tokens and language hidden states, discouraging over-pruning of vision-critical layers that facilitate cross-modal fusion. We optimize pruning plans with plan-level Group Relative Policy Optimization (GRPO) under a multi-objective return that combines task utility, hallucination robustness (POPE), compression, and a synaptic-flow-inspired stability proxy to reduce unproductive exploration in high-sparsity regimes. Experiments on LLaVA with POPE and ScienceQA demonstrate that HiPP-Prune discovers diverse non-dominated pruning plans and provides controllable robustness--utility trade-offs under matched sparsity budgets.

Related papers

• Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models [16.821897521676963]
  Vision-language models (VLMs) have achieved remarkable generalization across diverse multimodal tasks through large-scale pre-training.<n>We introduce a retraining-free VLM pruning framework called PORTA that derives a task- and modality-agnostic importance formulation.<n>We show that PORTA achieves competitive downstream performance under high sparsity without requiring any retraining, supporting efficient VLM compression.
  arXiv  Detail & Related papers  (2026-08-07T07:36:19Z)
• Not All Tokens See Equally: Perception-Grounded Policy Optimization for Large Vision-Language Models [38.47027398567909]
  Perception-Grounded Policy Optimization (PGPO) is a novel fine-grained credit assignment framework that dynamically reshapes advantages at the token level.<n>We show that PGPO actively amplifies learning signals for visually-dependent tokens while suppressing gradient noise from linguistic priors.<n>Both theoretical and empirical analyses confirm that PGPO effectively reduces gradient variance, prevents training collapse, and acts as a potent regularizer for robust, perception-grounded multimodal reasoning.
  arXiv  Detail & Related papers  (2026-04-02T09:53:20Z)
• ACPO: Counteracting Likelihood Displacement in Vision-Language Alignment with Asymmetric Constraints [26.309376587678354]
  We propose Asymmetric Constrained Preference Optimization (ACPO)<n>ACPO applies dynamic, target-oriented scaling to preference optimization.<n>We show that ACPO effectively reverses the chosen-reward degradation of standard DPO.
  arXiv  Detail & Related papers  (2026-03-23T16:26:11Z)
• Rethinking the Trust Region in LLM Reinforcement Learning [72.25890308541334]
  Proximal Policy Optimization (PPO) serves as the de facto standard algorithm for Large Language Models (LLMs)<n>We propose Divergence Proximal Policy Optimization (DPPO), which substitutes clipping with a more principled constraint.<n>DPPO achieves superior training and efficiency compared to existing methods, offering a more robust foundation for RL-based fine-tuning.
  arXiv  Detail & Related papers  (2026-02-04T18:59:04Z)
• 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)
• 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)
• PGPO: Enhancing Agent Reasoning via Pseudocode-style Planning Guided Preference Optimization [58.465778756331574]
  We propose a pseudocode-style Planning Guided Preference Optimization method called PGPO for effective agent learning.<n>With two planning-oriented rewards, PGPO further enhances LLM agents' ability to generate high-quality P-code Plans.<n>Experiments show that PGPO achieves superior performance on representative agent benchmarks and outperforms the current leading baselines.
  arXiv  Detail & Related papers  (2025-06-02T09:35:07Z)
• Reinforced Reasoning for Embodied Planning [18.40186665383579]
  Embodied planning requires agents to make coherent multi-step decisions based on dynamic visual observations and natural language goals.<n>We introduce a reinforcement fine-tuning framework that brings R1-style reasoning enhancement into embodied planning.
  arXiv  Detail & Related papers  (2025-05-28T07:21:37Z)
• REBEL: Reinforcement Learning via Regressing Relative Rewards [59.68420022466047]
  We propose REBEL, a minimalist RL algorithm for the era of generative models.<n>In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL.<n>We find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO.
  arXiv  Detail & Related papers  (2024-04-25T17:20:45Z)
• Modular Deep Reinforcement Learning for Continuous Motion Planning with Temporal Logic [59.94347858883343]
  This paper investigates the motion planning of autonomous dynamical systems modeled by Markov decision processes (MDP) The novelty is to design an embedded product MDP (EP-MDP) between the LDGBA and the MDP. The proposed LDGBA-based reward shaping and discounting schemes for the model-free reinforcement learning (RL) only depend on the EP-MDP states.
  arXiv  Detail & Related papers  (2021-02-24T01:11:25Z)
• Exploiting Submodular Value Functions For Scaling Up Active Perception [60.81276437097671]
  In active perception tasks, agent aims to select sensory actions that reduce uncertainty about one or more hidden variables. Partially observable Markov decision processes (POMDPs) provide a natural model for such problems. As the number of sensors available to the agent grows, the computational cost of POMDP planning grows exponentially.
  arXiv  Detail & Related papers  (2020-09-21T09:11:36Z)

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.