WorldPM: Scaling Human Preference Modeling

• URL: http://arxiv.org/abs/2505.10527v2
• Date: Sun, 18 May 2025 17:45:31 GMT
• Title: WorldPM: Scaling Human Preference Modeling
• Authors: Binghai Wang, Runji Lin, Keming Lu, Le Yu, Zhenru Zhang, Fei Huang, Chujie Zheng, Kai Dang, Yang Fan, Xingzhang Ren, An Yang, Binyuan Hui, Dayiheng Liu, Tao Gui, Qi Zhang, Xuanjing Huang, Yu-Gang Jiang, Bowen Yu, Jingren Zhou, Junyang Lin,
• Abstract summary: We propose World Preference Modeling$ (WorldPM) to emphasize this scaling potential.<n>We collect preference data from public forums covering diverse user communities.<n>We conduct extensive training using 15M-scale data across models ranging from 1.5B to 72B parameters.
• Score: 130.23230492612214
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Motivated by scaling laws in language modeling that demonstrate how test loss scales as a power law with model and dataset sizes, we find that similar laws exist in preference modeling. We propose World Preference Modeling$ (WorldPM) to emphasize this scaling potential, where World Preference embodies a unified representation of human preferences. In this paper, we collect preference data from public forums covering diverse user communities, and conduct extensive training using 15M-scale data across models ranging from 1.5B to 72B parameters. We observe distinct patterns across different evaluation metrics: (1) Adversarial metrics (ability to identify deceptive features) consistently scale up with increased training data and base model size; (2) Objective metrics (objective knowledge with well-defined answers) show emergent behavior in larger language models, highlighting WorldPM's scalability potential; (3) Subjective metrics (subjective preferences from a limited number of humans or AI) do not demonstrate scaling trends. Further experiments validate the effectiveness of WorldPM as a foundation for preference fine-tuning. Through evaluations on 7 benchmarks with 20 subtasks, we find that WorldPM broadly improves the generalization performance across human preference datasets of varying sizes (7K, 100K and 800K samples), with performance gains exceeding 5% on many key subtasks. Integrating WorldPM into our internal RLHF pipeline, we observe significant improvements on both in-house and public evaluation sets, with notable gains of 4% to 8% in our in-house evaluations.

Related papers

• WorldPrediction: A Benchmark for High-level World Modeling and Long-horizon Procedural Planning [52.36434784963598]
  We introduce WorldPrediction, a video-based benchmark for evaluating world modeling and procedural planning capabilities of different AI models.<n>We show that current frontier models barely achieve 57% accuracy on WorldPrediction-WM and 38% on WorldPrediction-PP whereas humans are able to solve both tasks perfectly.
  arXiv  Detail & Related papers  (2025-06-04T18:22:40Z)
• Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning [77.120955854093]
  We show that data diversity can be a strong predictor of generalization in language models.<n>We introduce G-Vendi, a metric that quantifies diversity via the entropy of model-induced gradients.<n>We present Prismatic Synthesis, a framework for generating diverse synthetic data.
  arXiv  Detail & Related papers  (2025-05-26T16:05:10Z)
• DataDecide: How to Predict Best Pretraining Data with Small Experiments [67.95896457895404]
  We release models, data, and evaluations in DataDecide -- the most extensive open suite of models over differences in data and scale.<n>We conduct controlled pretraining experiments across 25 corpora with differing sources, deduplication, and filtering up to 100B tokens, model sizes up to 1B parameters, and 3 random seeds.
  arXiv  Detail & Related papers  (2025-04-15T17:02:15Z)
• WorldModelBench: Judging Video Generation Models As World Models [57.776769550453594]
  Video generation models have rapidly progressed, positioning themselves as video world models capable of supporting decision-making applications like robotics and autonomous driving.<n>Current benchmarks fail to rigorously evaluate these claims, focusing only on general video quality.<n>We propose WorldModelBench, a benchmark designed to evaluate the world modeling capabilities of video generation models in application-driven domains.
  arXiv  Detail & Related papers  (2025-02-28T03:58:23Z)
• SMPLest-X: Ultimate Scaling for Expressive Human Pose and Shape Estimation [81.36747103102459]
  Expressive human pose and shape estimation (EHPS) unifies body, hands, and face motion capture with numerous applications.<n>Current state-of-the-art methods focus on training innovative architectural designs on confined datasets.<n>We investigate the impact of scaling up EHPS towards a family of generalist foundation models.
  arXiv  Detail & Related papers  (2025-01-16T18:59:46Z)
• PaPaGei: Open Foundation Models for Optical Physiological Signals [8.78925327256804]
  Photoplethysmography is the leading non-invasive technique for monitoring biosignals and cardiovascular health.<n>Machine learning models trained on PPG signals tend to be task-specific and struggle with generalization.<n>We present PaPaGei, the first open foundation model for PPG signals.
  arXiv  Detail & Related papers  (2024-10-27T18:18:06Z)
• Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment [51.14207112118503]
  We introduce preference embedding, an approach that embeds responses into a latent space to capture preferences efficiently.<n>We also propose preference score-based General Preference Optimization (GPO), which generalizes reward-based reinforcement learning from human feedback.<n>Our method may enhance the alignment of foundation models with nuanced human values.
  arXiv  Detail & Related papers  (2024-10-03T04:22:55Z)
• Scaling Laws Do Not Scale [54.72120385955072]
  Recent work has argued that as the size of a dataset increases, the performance of a model trained on that dataset will increase. We argue that this scaling law relationship depends on metrics used to measure performance that may not correspond with how different groups of people perceive the quality of models' output. Different communities may also have values in tension with each other, leading to difficult, potentially irreconcilable choices about metrics used for model evaluations.
  arXiv  Detail & Related papers  (2023-07-05T15:32:21Z)
• Understanding and Improving Model Averaging in Federated Learning on Heterogeneous Data [9.792805355704203]
  We study the loss landscape of model averaging in federated learning (FL) We decompose the expected loss of the global model into five factors related to the client models. We propose utilizing IMA on the global model at the late training phase to reduce its deviation from the expected speed.
  arXiv  Detail & Related papers  (2023-05-13T06:19:55Z)
• Prediction of SLAM ATE Using an Ensemble Learning Regression Model and 1-D Global Pooling of Data Characterization [3.4399698738841553]
  We introduce a novel method for predicting SLAM localization error based on the characterization of raw sensor inputs. The proposed method relies on using a random forest regression model trained on 1-D global pooled features that are generated from characterized raw sensor data. The paper also studies the impact of 12 different 1-D global pooling functions on regression quality, and the superiority of 1-D global averaging is quantitatively proven.
  arXiv  Detail & Related papers  (2023-03-01T16:12:47Z)

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.