Abstract:LLMs have demonstrated strong mathematical reasoning abilities by leveraging reinforcement learning with long chain-of-thought, yet they continue to struggle with theorem proving due to the lack of clear supervision signals when solely using natural language. Dedicated domain-specific languages like Lean provide clear supervision via formal verification of proofs, enabling effective training through reinforcement learning. In this work, we propose \textbf{Seed-Prover}, a lemma-style whole-proof reasoning model. Seed-Prover can iteratively refine its proof based on Lean feedback, proved lemmas, and self-summarization. To solve IMO-level contest problems, we design three test-time inference strategies that enable both deep and broad reasoning. Seed-Prover proves $78.1\%$ of formalized past IMO problems, saturates MiniF2F, and achieves over 50\% on PutnamBench, outperforming the previous state-of-the-art by a large margin. To address the lack of geometry support in Lean, we introduce a geometry reasoning engine \textbf{Seed-Geometry}, which outperforms previous formal geometry engines. We use these two systems to participate in IMO 2025 and fully prove 5 out of 6 problems. This work represents a significant advancement in automated mathematical reasoning, demonstrating the effectiveness of formal verification with long chain-of-thought reasoning.
Abstract:Simultaneous Interpretation (SI) represents one of the most daunting frontiers in the translation industry, with product-level automatic systems long plagued by intractable challenges: subpar transcription and translation quality, lack of real-time speech generation, multi-speaker confusion, and translated speech inflation, especially in long-form discourses. In this study, we introduce Seed-LiveInterpret 2.0, an end-to-end SI model that delivers high-fidelity, ultra-low-latency speech-to-speech generation with voice cloning capabilities. As a fully operational product-level solution, Seed-LiveInterpret 2.0 tackles these challenges head-on through our novel duplex speech-to-speech understanding-generating framework. Experimental results demonstrate that through large-scale pretraining and reinforcement learning, the model achieves a significantly better balance between translation accuracy and latency, validated by human interpreters to exceed 70% correctness in complex scenarios. Notably, Seed-LiveInterpret 2.0 outperforms commercial SI solutions by significant margins in translation quality, while slashing the average latency of cloned speech from nearly 10 seconds to a near-real-time 3 seconds, which is around a near 70% reduction that drastically enhances practical usability.
Abstract:Recently, test-time scaling Large Language Models (LLMs) have demonstrated exceptional reasoning capabilities across scientific and professional tasks by generating long chains-of-thought (CoT). As a crucial component for developing these reasoning models, reinforcement learning (RL), exemplified by Proximal Policy Optimization (PPO) and its variants, allows models to learn through trial and error. However, PPO can be time-consuming due to its inherent on-policy nature, which is further exacerbated by increasing response lengths. In this work, we propose Truncated Proximal Policy Optimization (T-PPO), a novel extension to PPO that improves training efficiency by streamlining policy update and length-restricted response generation. T-PPO mitigates the issue of low hardware utilization, an inherent drawback of fully synchronized long-generation procedures, where resources often sit idle during the waiting periods for complete rollouts. Our contributions are two-folds. First, we propose Extended Generalized Advantage Estimation (EGAE) for advantage estimation derived from incomplete responses while maintaining the integrity of policy learning. Second, we devise a computationally optimized mechanism that allows for the independent optimization of the policy and value models. By selectively filtering prompt and truncated tokens, this mechanism reduces redundant computations and accelerates the training process without sacrificing convergence performance. We demonstrate the effectiveness and efficacy of T-PPO on AIME 2024 with a 32B base model. The experimental results show that T-PPO improves the training efficiency of reasoning LLMs by up to 2.5x and outperforms its existing competitors.
Abstract:Objective: To characterize stigma dimensions, social, and related behavioral circumstances in people living with HIV (PLWHs) seeking care, using natural language processing methods applied to a large collection of electronic health record (EHR) clinical notes from a large integrated health system in the southeast United States. Methods: We identified 9,140 cohort of PLWHs from the UF Health IDR and performed topic modeling analysis using Latent Dirichlet Allocation (LDA) to uncover stigma dimensions, social, and related behavioral circumstances. Domain experts created a seed list of HIV-related stigma keywords, then applied a snowball strategy to iteratively review notes for additional terms until saturation was reached. To identify more target topics, we tested three keyword-based filtering strategies. Domain experts manually reviewed the detected topics using the prevalent terms and key discussion topics. Word frequency analysis was used to highlight the prevalent terms associated with each topic. In addition, we conducted topic variation analysis among subgroups to examine differences across age and sex-specific demographics. Results and Conclusion: Topic modeling on sentences containing at least one keyword uncovered a wide range of topic themes associated with HIV-related stigma, social, and related behaviors circumstances, including "Mental Health Concern and Stigma", "Social Support and Engagement", "Limited Healthcare Access and Severe Illness", "Treatment Refusal and Isolation" and so on. Topic variation analysis across age subgroups revealed differences. Extracting and understanding the HIV-related stigma dimensions, social, and related behavioral circumstances from EHR clinical notes enables scalable, time-efficient assessment, overcoming the limitations of traditional questionnaires and improving patient outcomes.
Abstract:To advance biomedical vison-language model capabilities through scaling up, fine-tuning, and instruction tuning, develop vision-language models with improved performance in handling long text, explore strategies to efficiently adopt vision language models for diverse multi-modal biomedical tasks, and examine the zero-shot learning performance. We developed two biomedical vision language models, BiomedGPT-Large and BiomedGPT-XLarge, based on an encoder-decoder-based transformer architecture. We fine-tuned the two models on 23 benchmark datasets from 6 multi-modal biomedical tasks including one image-only task (image classification), three language-only tasks (text understanding, text summarization and question answering), and two vision-language tasks (visual question answering and image captioning). We compared the developed scaled models with our previous BiomedGPT-Base model and existing prestigious models reported in the literature. We instruction-tuned the two models using a large-scale multi-modal biomedical instruction-tuning dataset and assessed the zero-shot learning performance and alignment accuracy.
Abstract:Model merging has emerged as a promising technique for enhancing large language models, though its application in large-scale pre-training remains relatively unexplored. In this paper, we present a comprehensive investigation of model merging techniques during the pre-training process. Through extensive experiments with both dense and Mixture-of-Experts (MoE) architectures ranging from millions to over 100 billion parameters, we demonstrate that merging checkpoints trained with constant learning rates not only achieves significant performance improvements but also enables accurate prediction of annealing behavior. These improvements lead to both more efficient model development and significantly lower training costs. Our detailed ablation studies on merging strategies and hyperparameters provide new insights into the underlying mechanisms while uncovering novel applications. Through comprehensive experimental analysis, we offer the open-source community practical pre-training guidelines for effective model merging.
Abstract:We present Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning. Seed1.5-VL is composed with a 532M-parameter vision encoder and a Mixture-of-Experts (MoE) LLM of 20B active parameters. Despite its relatively compact architecture, it delivers strong performance across a wide spectrum of public VLM benchmarks and internal evaluation suites, achieving the state-of-the-art performance on 38 out of 60 public benchmarks. Moreover, in agent-centric tasks such as GUI control and gameplay, Seed1.5-VL outperforms leading multimodal systems, including OpenAI CUA and Claude 3.7. Beyond visual and video understanding, it also demonstrates strong reasoning abilities, making it particularly effective for multimodal reasoning challenges such as visual puzzles. We believe these capabilities will empower broader applications across diverse tasks. In this report, we mainly provide a comprehensive review of our experiences in building Seed1.5-VL across model design, data construction, and training at various stages, hoping that this report can inspire further research. Seed1.5-VL is now accessible at https://www.volcengine.com/ (Volcano Engine Model ID: doubao-1-5-thinking-vision-pro-250428)
Abstract:Natural language generation (NLG) is the key technology to achieve generative artificial intelligence (AI). With the breakthroughs in large language models (LLMs), NLG has been widely used in various medical applications, demonstrating the potential to enhance clinical workflows, support clinical decision-making, and improve clinical documentation. Heterogeneous and diverse medical data modalities, such as medical text, images, and knowledge bases, are utilized in NLG. Researchers have proposed many generative models and applied them in a number of healthcare applications. There is a need for a comprehensive review of NLG methods and applications in the medical domain. In this study, we systematically reviewed 113 scientific publications from a total of 3,988 NLG-related articles identified using a literature search, focusing on data modality, model architecture, clinical applications, and evaluation methods. Following PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines, we categorize key methods, identify clinical applications, and assess their capabilities, limitations, and emerging challenges. This timely review covers the key NLG technologies and medical applications and provides valuable insights for future studies to leverage NLG to transform medical discovery and healthcare.
Abstract:We present VAPO, Value-based Augmented Proximal Policy Optimization framework for reasoning models., a novel framework tailored for reasoning models within the value-based paradigm. Benchmarked the AIME 2024 dataset, VAPO, built on the Qwen 32B pre-trained model, attains a state-of-the-art score of $\mathbf{60.4}$. In direct comparison under identical experimental settings, VAPO outperforms the previously reported results of DeepSeek-R1-Zero-Qwen-32B and DAPO by more than 10 points. The training process of VAPO stands out for its stability and efficiency. It reaches state-of-the-art performance within a mere 5,000 steps. Moreover, across multiple independent runs, no training crashes occur, underscoring its reliability. This research delves into long chain-of-thought (long-CoT) reasoning using a value-based reinforcement learning framework. We pinpoint three key challenges that plague value-based methods: value model bias, the presence of heterogeneous sequence lengths, and the sparsity of reward signals. Through systematic design, VAPO offers an integrated solution that effectively alleviates these challenges, enabling enhanced performance in long-CoT reasoning tasks.
Abstract:Reinforcement Learning from Human Feedback (RLHF) has emerged as a important paradigm for aligning large language models (LLMs) with human preferences during post-training. This framework typically involves two stages: first, training a reward model on human preference data, followed by optimizing the language model using reinforcement learning algorithms. However, current RLHF approaches may constrained by two limitations. First, existing RLHF frameworks often rely on Bradley-Terry models to assign scalar rewards based on pairwise comparisons of individual responses. However, this approach imposes significant challenges on reward model (RM), as the inherent variability in prompt-response pairs across different contexts demands robust calibration capabilities from the RM. Second, reward models are typically initialized from generative foundation models, such as pre-trained or supervised fine-tuned models, despite the fact that reward models perform discriminative tasks, creating a mismatch. This paper introduces Pairwise-RL, a RLHF framework that addresses these challenges through a combination of generative reward modeling and a pairwise proximal policy optimization (PPO) algorithm. Pairwise-RL unifies reward model training and its application during reinforcement learning within a consistent pairwise paradigm, leveraging generative modeling techniques to enhance reward model performance and score calibration. Experimental evaluations demonstrate that Pairwise-RL outperforms traditional RLHF frameworks across both internal evaluation datasets and standard public benchmarks, underscoring its effectiveness in improving alignment and model behavior.