Dual-Stage Value-Guided Inference with Margin-Based Reward Adjustment for Fast and Faithful VLM Captioning

• URL: http://arxiv.org/abs/2506.15649v1
• Date: Wed, 18 Jun 2025 17:23:36 GMT
• Title: Dual-Stage Value-Guided Inference with Margin-Based Reward Adjustment for Fast and Faithful VLM Captioning
• Authors: Ankan Deria, Adinath Madhavrao Dukre, Feilong Tang, Sara Atito, Sudipta Roy, Muhammad Awais, Muhammad Haris Khan, Imran Razzak,
• Abstract summary: We introduce textbfValue-guided Inference with Margin-based Reward (ViMaR), a two-stage inference framework that improves both efficiency and output fidelity.<n>ViMaR generates captions that are significantly more reliable, factually accurate, detailed, and explanatory, while achieving over 4$times$ speedup.
• Score: 23.851747078717473
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Despite significant advances in inference-time search for vision-language models (VLMs), existing approaches remain both computationally expensive and prone to unpenalized, low-confidence generations which often lead to persistent hallucinations. We introduce \textbf{Value-guided Inference with Margin-based Reward (ViMaR)}, a two-stage inference framework that improves both efficiency and output fidelity by combining a temporal-difference value model with a margin-aware reward adjustment. In the first stage, we perform a single pass to identify the highest-value caption among diverse candidates. In the second stage, we selectively refine only those segments that were overlooked or exhibit weak visual grounding, thereby eliminating frequently rewarded evaluations. A calibrated margin-based penalty discourages low-confidence continuations while preserving descriptive richness. Extensive experiments across multiple VLM architectures demonstrate that ViMaR generates captions that are significantly more reliable, factually accurate, detailed, and explanatory, while achieving over 4$\times$ speedup compared to existing value-guided methods. Specifically, we show that ViMaR trained solely on LLaVA Mistral-7B, \textit{generalizes effectively to guide decoding in a stronger unseen model}. To further validate this, we adapt the ViMaR to steer generation in LLaVA-OneVision-Qwen2-7B, leading to consistent improvements in caption quality and demonstrating robust cross-model guidance. This cross-model generalization highlights ViMaR's flexibility and modularity, positioning it as a scalable and transferable inference-time decoding strategy. Furthermore, when ViMaR-generated captions are used for self-training, the underlying models achieve substantial gains across a broad suite of visual comprehension benchmarks, underscoring the potential of fast, accurate, and self-improving VLM pipelines.

Related papers

• VLMQ: Efficient Post-Training Quantization for Large Vision-Language Models via Hessian Augmentation [8.891793681316992]
  Post-training quantization (PTQ) has emerged as an effective approach for compressing large models and accelerating their inference without retraining.<n>While PTQ has been extensively studied in the context of large language models (LLMs), its applicability to vision-language models (VLMs) remains underexplored.<n>We propose a novel importance-aware PTQ framework tailored for VLMs, dubbed VLMQ.
  arXiv  Detail & Related papers  (2025-08-05T11:57:03Z)
• MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings [75.0617088717528]
  MoCa is a framework for transforming pre-trained VLM backbones into effective bidirectional embedding models.<n>MoCa consistently improves performance across MMEB and ViDoRe-v2 benchmarks, achieving new state-of-the-art results.
  arXiv  Detail & Related papers  (2025-06-29T06:41:00Z)
• LaVi: Efficient Large Vision-Language Models via Internal Feature Modulation [17.318287255400175]
  We present LaVi, a novel LVLM that enables seamless and efficient vision-language fusion.<n>Unlike dominant LVLMs that rely on visual token concatenation, LaVi bypasses long-context expansion.<n>Compared to LLaVA-OV-7B, LaVi reduces FLOPs by 94.0%, improves inference speed by 3.1 times, and cuts memory usage in half.
  arXiv  Detail & Related papers  (2025-06-20T02:25:33Z)
• Autoregressive Semantic Visual Reconstruction Helps VLMs Understand Better [44.15671594378141]
  We introduce Autoregressive Semantic Visual Reconstruction (ASVR), which enables joint learning of visual and textual modalities within a unified autoregressive framework.<n>ASVR improves LLaVA-1.5 by 5% in average scores across 14 multimodal benchmarks.
  arXiv  Detail & Related papers  (2025-06-10T17:57:50Z)
• Reinforcing Multimodal Understanding and Generation with Dual Self-rewards [56.08202047680044]
  Large language models (LLMs) unify cross-model understanding and generation into a single framework.<n>Current solutions require external supervision (e.g., human feedback or reward models) and only address unidirectional tasks.<n>We introduce a self-supervised dual reward mechanism to reinforce the understanding and generation capabilities of LMMs.
  arXiv  Detail & Related papers  (2025-06-09T17:38:45Z)
• From Captions to Rewards (CAREVL): Leveraging Large Language Model Experts for Enhanced Reward Modeling in Large Vision-Language Models [58.16075709485292]
  CAREVL is a novel method for preference reward modeling by reliably using both high- and low-confidence data.<n> CAREVL achieves performance improvements over traditional distillation-based methods on VL-RewardBench and MLLM-as-a-Judge benchmark.
  arXiv  Detail & Related papers  (2025-03-08T16:13:18Z)
• Scaling Inference-Time Search with Vision Value Model for Improved Visual Comprehension [95.63899307791665]
  Vision Value Model (VisVM) can guide VLM inference-time search to generate responses with better visual comprehension.<n>In this paper, we present VisVM that can guide VLM inference-time search to generate responses with better visual comprehension.
  arXiv  Detail & Related papers  (2024-12-04T20:35:07Z)
• Fine-Grained Verifiers: Preference Modeling as Next-token Prediction in Vision-Language Alignment [57.0121616203175]
  We propose FiSAO, a novel self-alignment method that utilizes the model's own visual encoder as a fine-grained verifier to improve vision-language alignment.<n>By leveraging token-level feedback from the vision encoder, FiSAO significantly improves vision-language alignment, even surpassing traditional preference tuning methods that require additional data.
  arXiv  Detail & Related papers  (2024-10-18T03:34:32Z)
• Mono-InternVL: Pushing the Boundaries of Monolithic Multimodal Large Language Models with Endogenous Visual Pre-training [48.455597568212944]
  We present Mono-InternVL, a novel monolithic MLLM that seamlessly integrates a set of visual experts via a multimodal mixture-of-experts structure.<n>In particular, EViP is designed as a progressive learning process for visual experts, which aims to fully exploit the visual knowledge from noisy data to high-quality data.
  arXiv  Detail & Related papers  (2024-10-10T17:59:22Z)
• Mitigating Object Hallucination in Large Vision-Language Models via Image-Grounded Guidance [51.30560006045442]
  Image-gRounded guIdaNcE (MARINE) is a framework that is both training-free and API-free.<n>MARINE effectively and efficiently reduces object hallucinations during inference by introducing image-grounded guidance to LVLMs.<n>Our framework's flexibility further allows for the integration of multiple vision models, enabling more reliable and robust object-level guidance.
  arXiv  Detail & Related papers  (2024-02-13T18:59:05Z)

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.