Close Menu
    DevStackTipsDevStackTips
    • Home
    • News & Updates
      1. Tech & Work
      2. View All

      Why Non-Native Content Designers Improve Global UX

      July 18, 2025

      DevOps won’t scale without platform engineering and here’s why your teams are still stuck

      July 18, 2025

      This week in AI dev tools: Slack’s enterprise search, Claude Code’s analytics dashboard, and more (July 18, 2025)

      July 18, 2025

      Report: 71% of tech leaders won’t hire devs without AI skills

      July 17, 2025

      Remedy offers update on ‘FBC: Firebreak,’ details coming improvements — “We’ve seen many players come into the game and leave within the first hour.”

      July 18, 2025

      I ran with Samsung’s Galaxy Watch 8 Classic, and it both humbled and motivated me

      July 18, 2025

      You can finally move Chrome’s address bar on Android – here’s how

      July 18, 2025

      Is your Ring camera showing strange logins? Here’s what’s going on

      July 18, 2025
    • Development
      1. Algorithms & Data Structures
      2. Artificial Intelligence
      3. Back-End Development
      4. Databases
      5. Front-End Development
      6. Libraries & Frameworks
      7. Machine Learning
      8. Security
      9. Software Engineering
      10. Tools & IDEs
      11. Web Design
      12. Web Development
      13. Web Security
      14. Programming Languages
        • PHP
        • JavaScript
      Featured

      The details of TC39’s last meeting

      July 18, 2025
      Recent

      The details of TC39’s last meeting

      July 18, 2025

      Online Examination System using PHP and MySQL

      July 18, 2025

      A tricky, educational quiz: it’s about time..

      July 18, 2025
    • Operating Systems
      1. Windows
      2. Linux
      3. macOS
      Featured

      Remedy offers update on ‘FBC: Firebreak,’ details coming improvements — “We’ve seen many players come into the game and leave within the first hour.”

      July 18, 2025
      Recent

      Remedy offers update on ‘FBC: Firebreak,’ details coming improvements — “We’ve seen many players come into the game and leave within the first hour.”

      July 18, 2025

      Ubuntu 25.10 Shrinks its Raspberry Pi Install Footprint

      July 18, 2025

      Microsoft kills Movies & TV storefront on Windows and Xbox — here’s what will happen to your purchased media

      July 18, 2025
    • Learning Resources
      • Books
      • Cheatsheets
      • Tutorials & Guides
    Home»Development»Machine Learning»ByteDance Introduces VAPO: A Novel Reinforcement Learning Framework for Advanced Reasoning Tasks

    ByteDance Introduces VAPO: A Novel Reinforcement Learning Framework for Advanced Reasoning Tasks

    April 10, 2025
    ByteDance Introduces VAPO: A Novel Reinforcement Learning Framework for Advanced Reasoning Tasks

    In the Large Language Models (LLM) RL training, value-free methods like GRPO and DAPO have shown great effectiveness. The true potential lies in value-based methods, which allow more precise credit assignment by accurately tracing each action’s impact on subsequent returns. This precision is crucial for complex reasoning, where subtle errors can lead to catastrophic failures. However, training effective value models for long chain-of-thought (CoT) tasks face challenges: achieving low bias despite lengthy trajectories, managing distinct preferences of short and long responses, and addressing reward signal sparsity. Despite their theoretical advantages, these difficulties have hindered the full realization of value-based methods.

    Value-based reinforcement learning methods for LLMs face three significant challenges when applied to long chain-of-thought reasoning tasks. First, the Value Model Bias issue identified in VC-PPO shows that initializing value models with reward models introduces positive bias. Second, Heterogeneous Sequence Lengths in complex reasoning tasks create difficulties for standard approaches like GAE with fixed parameters, which cannot effectively adapt to sequences ranging from very short to extremely long. Third, the Sparsity of the Reward Signal becomes problematic in verifier-based tasks that provide binary feedback rather than continuous values. This sparsity is worsened by lengthy CoT responses, creating a difficult exploration-exploitation trade-off during optimization.

    Researchers from ByteDance Seed have proposed Value Augmented Proximal Policy Optimization (VAPO), a value-based RL training framework to address the challenges of long CoT reasoning tasks. VAPO introduces three key innovations: a detailed value-based training framework with superior performance and efficiency, a Length-adaptive GAE mechanism that adjusts the parameter based on response lengths to optimize advantage estimation, and a systematic integration of techniques from prior research. VAPO combines these components to create a system where the collective improvements exceed what individual enhancements could achieve independently.  Using the Qwen2.5-32B model without SFT data, VAPO improves scores from 5 to 60, surpassing previous state-of-the-art methods by 10 points.

    The VAPO is built upon the PPO algorithm with several key modifications to enhance mathematical reasoning capabilities. Training dynamics analysis reveals VAPO’s superior characteristics compared to DAPO, including smoother training curves indicating more stable optimization, better length scaling which enhances generalization capabilities, faster score growth due to the granular signals provided by the value model, and lower entropy in later training stages. While reduced entropy could potentially limit exploration, the method balances this trade-off effectively, resulting in minimal performance impact while improving reproducibility and stability. This shows how VAPO’s decisions directly address the core challenges of value-based RL in complex reasoning tasks.

    While DeepSeek R1 using GRPO achieves 47 points on AIME24 and DAPO reaches 50 points, VAPO matches DAPO’s performance on Qwen-32b with just 60% of the update steps and achieves a new state-of-the-art score of 60.4 within only 5,000 steps. Vanilla PPO achieves only 5 points due to value model learning collapse, but VAPO finally achieves 60 points. Ablation studies validated the effectiveness of the seven proposed modifications: Value-Pretraining prevents collapse, decoupled GAE enables full optimization of long-form responses, adaptive GAE balances short and long response optimization, Clip-higher encourages thorough exploration, Token-level loss increases long response weighting, positive-example LM loss adds 6 points, and Group-Sampling contributes 5 points to the final performance.

    In this paper, researchers introduced VAPO, an algorithm that utilizes the Qwen2.5-32B model to achieve state-of-the-art performance on the AIME24 benchmark. By introducing seven innovative techniques on top of the PPO framework, VAPO significantly refines value learning and creates an optimal balance between exploration and exploitation. This value-based approach decisively outperforms value-free methods like GRPO and DAPO, establishing a new performance ceiling for reasoning tasks. It addresses fundamental challenges in training value models for long CoT scenarios, providing a robust foundation for advancing LLMs in reasoning-intensive applications.


    Check out the Paper. All credit for this research goes to the researchers of this project. Also, feel free to follow us on Twitter and don’t forget to join our 85k+ ML SubReddit.

    🔥 [Register Now] miniCON Virtual Conference on OPEN SOURCE AI: FREE REGISTRATION + Certificate of Attendance + 3 Hour Short Event (April 12, 9 am- 12 pm PST) + Hands on Workshop [Sponsored]

    The post ByteDance Introduces VAPO: A Novel Reinforcement Learning Framework for Advanced Reasoning Tasks appeared first on MarkTechPost.

    Source: Read More 

    Facebook Twitter Reddit Email Copy Link
    Previous ArticleAutomating regulatory compliance: A multi-agent solution using Amazon Bedrock and CrewAI
    Next Article T* and LV-Haystack: A Spatially-Guided Temporal Search Framework for Efficient Long-Form Video Understanding

    Related Posts

    Machine Learning

    How to Evaluate Jailbreak Methods: A Case Study with the StrongREJECT Benchmark

    July 18, 2025
    Machine Learning

    Manage multi-tenant Amazon Bedrock costs using application inference profiles

    July 18, 2025
    Leave A Reply Cancel Reply

    For security, use of Google's reCAPTCHA service is required which is subject to the Google Privacy Policy and Terms of Use.

    Continue Reading

    I played Capcom’s Pragmata, and it’s not what you expect

    News & Updates

    Perficient Wins the Gold: Globee® Customer Excellence Award for Customer Success Story 

    Development

    phởdav is a minimal WebDAV server

    Linux

    Microsoft Addresses Entra ID Token Logging Issue, Alerts to Protect Users

    Security

    Highlights

    I never thought I’d praise a kickstand power bank – until I tried this one

    June 30, 2025

    The ESR Qi2 MagSafe battery pack is the first kickstand power bank that I’ve used…

    CVE-2025-5230 – PHPGurukul Online Nurse Hiring System SQL Injection Vulnerability

    May 27, 2025

    Mastering Your Smartphone: 10 Features You Didn’t Know Existed

    June 22, 2025

    CVE-2025-4229 – Palo Alto Networks PAN-OS SD-WAN Information Disclosure Vulnerability

    June 13, 2025
    © DevStackTips 2025. All rights reserved.
    • Contact
    • Privacy Policy

    Type above and press Enter to search. Press Esc to cancel.