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

      AI and its impact on the developer experience, or ‘where is the joy?’

      July 23, 2025

      Google launches OSS Rebuild tool to improve trust in open source packages

      July 23, 2025

      AI-enabled software development: Risk of skill erosion or catalyst for growth?

      July 23, 2025

      BrowserStack launches Figma plugin for detecting accessibility issues in design phase

      July 22, 2025

      Power bank slapped with a recall? Stop using it now – here’s why

      July 23, 2025

      I recommend these budget earbuds over pricier Bose and Sony models – here’s why

      July 23, 2025

      Microsoft’s big AI update for Windows 11 is here – what’s new

      July 23, 2025

      Slow internet speed on Linux? This 30-second fix makes all the difference

      July 23, 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

      Singleton and Scoped Container Attributes in Laravel 12.21

      July 23, 2025
      Recent

      Singleton and Scoped Container Attributes in Laravel 12.21

      July 23, 2025

      wulfheart/laravel-actions-ide-helper

      July 23, 2025

      lanos/laravel-cashier-stripe-connect

      July 23, 2025
    • Operating Systems
      1. Windows
      2. Linux
      3. macOS
      Featured

      ‘Wuchang: Fallen Feathers’ came close to fully breaking me multiple times — a soulslike as brutal and as beautiful as it gets

      July 23, 2025
      Recent

      ‘Wuchang: Fallen Feathers’ came close to fully breaking me multiple times — a soulslike as brutal and as beautiful as it gets

      July 23, 2025

      Sam Altman is “terrified” of voice ID fraudsters embracing AI — and threats of US bioweapon attacks keep him up at night

      July 23, 2025

      NVIDIA boasts a staggering $111 million in market value per employee — since it became the world’s first $4 trillion company

      July 23, 2025
    • Learning Resources
      • Books
      • Cheatsheets
      • Tutorials & Guides
    Home»Development»Machine Learning»NVIDIA Introduces CLIMB: A Framework for Iterative Data Mixture Optimization in Language Model Pretraining

    NVIDIA Introduces CLIMB: A Framework for Iterative Data Mixture Optimization in Language Model Pretraining

    April 19, 2025

    Challenges in Constructing Effective Pretraining Data Mixtures

    As large language models (LLMs) scale in size and capability, the choice of pretraining data remains a critical determinant of downstream performance. Most LLMs are trained on large, web-scale datasets such as Common Crawl, which provide broad coverage but lack explicit domain labels. This introduces difficulties in curating mixtures that balance general knowledge with domain-specific expertise.

    Manual dataset curation, as seen in efforts like The Pile, is labor-intensive and does not scale well. Moreover, the nonlinear relationship between data composition and model performance makes it non-trivial to determine what proportions of domain data are optimal. These constraints motivate the need for automated, scalable, and adaptive data selection methods.

    CLIMB: An Iterative Framework for Data Mixture Discovery

    To address this, NVIDIA researchers propose CLIMB—CLustering-based Iterative Data Mixture Bootstrapping—a framework that automates the discovery and refinement of data mixtures for language model pretraining. CLIMB combines unsupervised clustering with iterative optimization to identify mixtures that are well-suited for general or domain-specific objectives.

    The pipeline begins by embedding large-scale text data into a semantic space using pretrained encoders. K-means clustering is then applied to organize the data into coherent groups, which are pruned and merged based on content quality and redundancy. This forms the basis for constructing candidate mixtures.

    Subsequently, CLIMB uses proxy models to evaluate sampled mixtures and fits a regression-based predictor (e.g., LightGBM) to estimate mixture performance. An iterative bootstrapping procedure progressively refines the sampling space, prioritizing high-performing configurations. This allows CLIMB to converge on an effective data mixture under a fixed compute budget.

    Technical Details and Design Considerations

    The optimization process is framed as a bi-level problem: at the lower level, proxy models are trained on candidate mixtures; at the upper level, a predictor is learned to approximate performance outcomes. This predictor guides further sampling and pruning, enabling efficient exploration of the mixture space.

    CLIMB supports sparsity in mixture weights, encouraging the discovery of compact, domain-relevant data subsets. The use of clustering over embeddings—rather than token-level features—ensures semantic coherence within clusters. The iterative refinement is structured to balance breadth (search space coverage) with depth (predictive accuracy), and ablation studies confirm that careful compute allocation across iterations improves convergence and final performance.

    The framework also exhibits robustness across proxy model sizes and cluster granularities. While larger proxy models yield slightly better predictions, even smaller models preserve key structural trends. Similarly, CLIMB is relatively insensitive to initial cluster count, provided it is within a reasonable range.

    Empirical Evaluation and Observations

    CLIMB was evaluated on several general reasoning tasks, including PIQA, ARC (Easy and Challenge), HellaSwag, and WinoGrande. A 1B-parameter model trained on CLIMB-discovered mixtures achieved an average accuracy of 60.41%, outperforming comparable baselines such as DoReMi and RegMix.

    When extended to 400B-token pretraining, this 1B model outperformed Llama-3.2-1B by 2.0% on a broad suite of benchmarks. Similarly, in the sub-500M model category, CLIMB-based pretraining led to consistent improvements over models like SmolLM and TinyLlama.

    Domain specialization further highlights CLIMB’s utility. In targeted MMLU benchmarks across STEM, humanities, and social sciences, CLIMB-trained models outperformed both random selection and exhaustive search baselines. The iterative process showed consistent gains over each stage, indicating effective guidance from the predictive model.

    To facilitate reproducibility and further research, NVIDIA has released two resources:

    • ClimbLab: A 1.2-trillion-token corpus organized into 20 semantic clusters.
    • ClimbMix: A 400-billion-token optimized mixture for efficient pretraining.

    Models trained on ClimbMix outperform those trained on datasets like Nemotron-CC and SmolLM under equivalent token budgets, demonstrating improved scaling characteristics.

    Conclusion

    CLIMB presents a systematic approach for optimizing data mixtures in LLM pretraining. By combining semantic clustering with proxy-based iterative search, it avoids reliance on manual annotations or static heuristics. The method supports both generalist and specialist training goals and adapts to varying compute and data constraints.

    This framework contributes to ongoing efforts in data-centric AI by offering a scalable and principled alternative to handcrafted data pipelines. Its empirical performance underscores the importance of data mixture optimization in maximizing model utility, particularly under fixed resource budgets.


    Check out the Paper, ClimbLab on HF and ClimbMix on HF . Also, don’t forget to follow us on Twitter and join our Telegram Channel and LinkedIn Group. Don’t Forget to join our 90k+ ML SubReddit.

    🔥 [Register Now] miniCON Virtual Conference on AGENTIC AI: FREE REGISTRATION + Certificate of Attendance + 4 Hour Short Event (May 21, 9 am- 1 pm PST) + Hands on Workshop

    The post NVIDIA Introduces CLIMB: A Framework for Iterative Data Mixture Optimization in Language Model Pretraining appeared first on MarkTechPost.

    Source: Read More 

    Facebook Twitter Reddit Email Copy Link
    Previous ArticleSelenium Report Generation: A Detailed Analysis
    Next Article OpenAI Releases a Technical Playbook for Enterprise AI Integration

    Related Posts

    Machine Learning

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

    July 23, 2025
    Machine Learning

    FastVLM: Efficient Vision Encoding for Vision Language Models

    July 23, 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

    CVE-2024-51099 – PHPGURUKUL Medical Card Generation System Reflected XSS

    Common Vulnerabilities and Exposures (CVEs)

    Go VIRAL on YouTube in 48 Hours: Srinidhi Ranganathan and BookSpotz Will Explode Your Product to Fame with 10 Lakhs Rupees!

    Artificial Intelligence

    FOSS Weekly #25.25: Nitrux Hyprland, Joplin Tips, Denmark Ditching Microsoft, Tiling Moves and More Linux Stuff

    Linux

    CVE-2025-1330 – IBM CICS TX DNS Code Injection

    Common Vulnerabilities and Exposures (CVEs)

    Highlights

    CVE-2025-0993 – GitLab Denial of Service

    May 22, 2025

    CVE ID : CVE-2025-0993

    Published : May 22, 2025, 3:16 p.m. | 1 hour, 31 minutes ago

    Description : An issue has been discovered in GitLab CE/EE affecting all versions before 17.10.7, 17.11 before 17.11.3, and 18.0 before 18.0.1. This could allow an authenticated attacker to cause a denial of service condition by exhausting server resources.

    Severity: 7.5 | HIGH

    Visit the link for more details, such as CVSS details, affected products, timeline, and more…

    CVE-2025-6578 – “Code-projects Simple Online Hotel Reservation System SQL Injection Vulnerability”

    June 24, 2025

    z/VSE: The forgotten operating system

    May 22, 2025

    Microsoft shadow launched this mouse that pairs with the new Surface Pro and Surface Laptop perfectly

    May 8, 2025
    © DevStackTips 2025. All rights reserved.
    • Contact
    • Privacy Policy

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