DreamPRM-1.5: Unlocking the Potential of Each Instance for Multimodal Process Reward Model Training

• URL: http://arxiv.org/abs/2509.05542v2
• Date: Tue, 21 Oct 2025 09:47:56 GMT
• Title: DreamPRM-1.5: Unlocking the Potential of Each Instance for Multimodal Process Reward Model Training
• Authors: Qi Cao, Pengtao Xie,
• Abstract summary: DreamPRM-1.5 is an instance-level reweighting framework that assigns an adaptive weight to every training example via bi-level optimization.<n>It attains 84.6 accuracy on the MMMU validation set, 31.3 accuracy on R-Bench-V and, when paired with a leading backbone, achieves first-place results on public multimodal reasoning leaderboards.
• Score: 28.02129783121819
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Training multimodal process reward models (PRMs) is hard due to (i) distribution shift between training set and test set and (ii) quality imbalance across training data samples. While domain-level reweighting (e.g., DreamPRM) aligns training with test-time objectives, it leaves a clear gap to an oracle upper bound (pass@N), even under a "sanity check" that uses test set data to probe headroom -- pointing to meta-level under-parameterization. We introduce DreamPRM-1.5, an instance-level reweighting framework that assigns an adaptive weight to every training example via bi-level optimization. To realize instance reweighting across scales, we develop two complementary regimes: Instance Table, which learns explicit per-sample weights and excels on small/medium data, and Instance Net, a lightweight neural network that generalizes better and scales to large corpora. A practical, stable training recipe -- time-scale matching between upper/lower updates, cold-start initialization, and bounded-range weights -- prevents divergence. Integrated with test-time scaling, DreamPRM-1.5 attains 84.6 accuracy on the MMMU validation set, 31.3 accuracy on R-Bench-V and, when paired with a leading backbone (e.g., GPT-5-mini), achieves first-place results on public multimodal reasoning leaderboards. Moreover, extensive experiments, including benchmark evaluations, baseline comparisons, and a sanity check, demonstrate that DreamPRM-1.5 closes the gap toward the oracle, achieves leading performance, and trains stably.

Related papers

• Bagging-Based Model Merging for Robust General Text Embeddings [73.51674133699196]
  General-purpose text embedding models underpin a wide range of NLP and information retrieval applications.<n>We present a systematic study of multi-task training for text embeddings from two perspectives: data scheduling and model merging.<n>We propose Bagging-based rObust mOdel Merging (BOOM), which trains multiple embedding models on sampled subsets and merges them into a single model.
  arXiv  Detail & Related papers  (2026-02-05T15:45:08Z)
• Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models [49.911784762244814]
  TraceRL is a trajectory-aware reinforcement learning framework for diffusion language models (DLMs)<n>We derive a series of state-of-the-art diffusion language models, namely TraDo.<n>TraDo-8B-Instruct achieves relative accuracy improvements of 6.1% over Qwen2.5-7B-Instruct and 51.3% over Llama3.1-8B-Instruct on mathematical reasoning benchmarks.
  arXiv  Detail & Related papers  (2025-09-08T17:58:06Z)
• MiniCPM4: Ultra-Efficient LLMs on End Devices [126.22958722174583]
  MiniCPM4 is a highly efficient large language model (LLM) designed explicitly for end-side devices.<n>We achieve this efficiency through systematic innovation in four key dimensions: model architecture, training data, training algorithms, and inference systems.
  arXiv  Detail & Related papers  (2025-06-09T16:16:50Z)
• AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners [35.730489673985936]
  Self-Taughters (STaR) is an integral part of the training pipeline of self-improving reasoning Language Models (LMs)<n>We introduce Adaptive STaR (AdaSTaR), a novel algorithm that rectifies this by integrating two adaptive sampling principles.<n>AdaSTaR achieves best test accuracy in all instances and reduces training FLOPs by an average of 58.6% against an extensive list of baselines.
  arXiv  Detail & Related papers  (2025-05-22T07:24:11Z)
• ToReMi: Topic-Aware Data Reweighting for Dynamic Pre-Training Data Selection [28.75333303894706]
  ToReMi is a novel framework that adjusts training sample weights according to their topical associations and observed learning patterns.<n>Our experiments reveal that ToReMi variants consistently achieve superior performance over conventional pre-training approaches.
  arXiv  Detail & Related papers  (2025-04-01T12:06:42Z)
• Test-Time Alignment via Hypothesis Reweighting [56.71167047381817]
  Large pretrained models often struggle with underspecified tasks.<n>We propose a novel framework to address the challenge of aligning models to test-time user intent.
  arXiv  Detail & Related papers  (2024-12-11T23:02:26Z)
• Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization [65.64108848398696]
  We introduce a preference optimization (PO) process to enhance the multimodal reasoning capabilities of MLLMs.<n>Specifically, we design an automated preference data construction pipeline to create MMPR, a high-quality, large-scale multimodal reasoning preference dataset.<n>We explore integrating PO with MLLMs, developing a simple yet effective method, termed Mixed Preference Optimization (MPO), which boosts multimodal CoT performance.
  arXiv  Detail & Related papers  (2024-11-15T18:59:27Z)
• Leveraging Foundation Models for Multi-modal Federated Learning with Incomplete Modality [41.79433449873368]
  We propose a novel multi-modal federated learning method, Federated Multi-modal contrastiVe training with Pre-trained completion (FedMVP) FedMVP integrates the large-scale pre-trained models to enhance the federated training. We demonstrate that the model achieves superior performance over two real-world image-text classification datasets.
  arXiv  Detail & Related papers  (2024-06-16T19:18:06Z)
• An Approach to Build Zero-Shot Slot-Filling System for Industry-Grade Conversational Assistants [9.537527104259153]
  Key requirements of this system include: 1) usage of smaller-sized models to meet low latency requirements and to enable convenient and cost-effective cloud and customer premise deployments. We adopt a fine-tuning approach where a pre-trained LLM is fine-tuned into a slot-filling model using task specific data. Results show that our prescribed approach for slot-filling model building has resulted in 6.9% relative improvement of F1 metric over the best baseline on a realistic benchmark, while at the same time reducing the latency by 57%.
  arXiv  Detail & Related papers  (2024-06-13T06:24:52Z)
• Take the Bull by the Horns: Hard Sample-Reweighted Continual Training Improves LLM Generalization [165.98557106089777]
  A key challenge is to enhance the capabilities of large language models (LLMs) amid a looming shortage of high-quality training data. Our study starts from an empirical strategy for the light continual training of LLMs using their original pre-training data sets. We then formalize this strategy into a principled framework of Instance-Reweighted Distributionally Robust Optimization.
  arXiv  Detail & Related papers  (2024-02-22T04:10:57Z)
• Sample Weight Estimation Using Meta-Updates for Online Continual Learning [7.832189413179361]
  Online Meta-learning for Sample Importance (OMSI) strategy approximates sample weights for a mini-batch in an online CL stream. OMSI enhances both learning and retained accuracy in a controlled noisy-labeled data stream.
  arXiv  Detail & Related papers  (2024-01-29T09:04:45Z)
• G-Meta: Distributed Meta Learning in GPU Clusters for Large-Scale Recommender Systems [16.343248795178685]
  This paper provides a framework for large-scale training for optimization-based Meta DLRM models over the textbfGPU cluster. Various experimental results show that G-Meta achieves notable training speed without loss of statistical performance.
  arXiv  Detail & Related papers  (2024-01-09T03:35:43Z)
• When Parameter-efficient Tuning Meets General-purpose Vision-language Models [65.19127815275307]
  PETAL revolutionizes the training process by requiring only 0.5% of the total parameters, achieved through a unique mode approximation technique. Our experiments reveal that PETAL not only outperforms current state-of-the-art methods in most scenarios but also surpasses full fine-tuning models in effectiveness.
  arXiv  Detail & Related papers  (2023-12-16T17:13:08Z)
• Learning to Re-weight Examples with Optimal Transport for Imbalanced Classification [74.62203971625173]
  Imbalanced data pose challenges for deep learning based classification models. One of the most widely-used approaches for tackling imbalanced data is re-weighting. We propose a novel re-weighting method based on optimal transport (OT) from a distributional point of view.
  arXiv  Detail & Related papers  (2022-08-05T01:23:54Z)
• Distributionally Robust Models with Parametric Likelihood Ratios [123.05074253513935]
  Three simple ideas allow us to train models with DRO using a broader class of parametric likelihood ratios. We find that models trained with the resulting parametric adversaries are consistently more robust to subpopulation shifts when compared to other DRO approaches.
  arXiv  Detail & Related papers  (2022-04-13T12:43:12Z)
• Adaptive Consistency Regularization for Semi-Supervised Transfer Learning [31.66745229673066]
  We consider semi-supervised learning and transfer learning jointly, leading to a more practical and competitive paradigm. To better exploit the value of both pre-trained weights and unlabeled target examples, we introduce adaptive consistency regularization. Our proposed adaptive consistency regularization outperforms state-of-the-art semi-supervised learning techniques such as Pseudo Label, Mean Teacher, and MixMatch.
  arXiv  Detail & Related papers  (2021-03-03T05:46:39Z)
• Meta-Learned Confidence for Few-shot Learning [60.6086305523402]
  A popular transductive inference technique for few-shot metric-based approaches, is to update the prototype of each class with the mean of the most confident query examples. We propose to meta-learn the confidence for each query sample, to assign optimal weights to unlabeled queries. We validate our few-shot learning model with meta-learned confidence on four benchmark datasets.
  arXiv  Detail & Related papers  (2020-02-27T10:22:17Z)

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.