Dan Alistarh

Venues

Frequent coauthors

Papers

1. Speculative Decoding Speed-of-Light: Optimal Lower Bounds via Branching Random Walks
   EACL 2026 ·

   Sergey Pankratov

   DBLP profile ↗ORCID search ↗
2. "Give Me BF16 or Give Me Death"? Accuracy-Performance Trade-Offs in LLM Quantization
   ACL 2025 ·

   Eldar Kurtic

   DBLP profile ↗ORCID search ↗
3. Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models
   ICML 2025 ·

   Alina Shutova

   DBLP profile ↗ORCID search ↗
4. Efficient Data Selection at Scale via Influence Distillation
   NeurIPS 2025 ·

   Mahdi Nikdan

   DBLP profile ↗ORCID search ↗
5. EvoPress: Accurate Dynamic Model Compression via Evolutionary Search
   ICML 2025 ·

   Oliver Sieberling

   DBLP profile ↗ORCID search ↗
6. HALO: Hadamard-Assisted Lower-Precision Optimization for LLMs
   NeurIPS 2025 ·

   Saleh Ashkboos

   DBLP profile ↗ORCID search ↗
7. HIGGS: Pushing the Limits of Large Language Model Quantization via the Linearity Theorem
   NAACL 2025 ·

   Vladimir Malinovskii

   DBLP profile ↗ORCID search ↗
8. Hogwild! Inference: Parallel LLM Generation via Concurrent Attention
   NeurIPS 2025 ·

   Gleb Rodionov

   DBLP profile ↗ORCID search ↗
9. Hybrid Decentralized Optimization: Leveraging Both First- and Zeroth-Order Optimizers for Faster Convergence
   AAAI 2025 ·

   Shayan Talaei

   DBLP profile ↗ORCID search ↗
10. LDAdam: Adaptive Optimization from Low-Dimensional Gradient Statistics
    ICLR 2025 ·

    Thomas Robert

    DBLP profile ↗ORCID search ↗
11. Layer-wise Quantization for Quantized Optimistic Dual Averaging
    ICML 2025 ·

    Anh Duc Nguyen

    DBLP profile ↗ORCID search ↗
12. QuEST: Stable Training of LLMs with 1-Bit Weights and Activations
    ICML 2025 ·

    Andrei Panferov

    DBLP profile ↗ORCID search ↗
13. Quartet: Native FP4 Training Can Be Optimal for Large Language Models
    NeurIPS 2025 ·

    Roberto L. Castro

    DBLP profile ↗ORCID search ↗
14. Scalable Mechanistic Neural Networks
    ICLR 2025 ·

    Jiale Chen

    DBLP profile ↗ORCID search ↗
15. The Journey Matters: Average Parameter Count over Pre-training Unifies Sparse and Dense Scaling Laws
    ICLR 2025 ·

    Tian Jin

    DBLP profile ↗ORCID search ↗
16. Unified Scaling Laws for Compressed Representations
    NeurIPS 2025 ·

    Andrei Panferov

    DBLP profile ↗ORCID search ↗
17. Wasserstein Distances, Neuronal Entanglement, and Sparsity
    ICLR 2025 ·

    Shashata Sawmya

    DBLP profile ↗ORCID search ↗
18. AsGrad: A Sharp Unified Analysis of Asynchronous-SGD Algorithms
    AISTATS 2024 ·

    Rustem Islamov

    DBLP profile ↗ORCID search ↗
19. Communication-Efficient Federated Learning With Data and Client Heterogeneity
    AISTATS 2024 ·

    Hossein Zakerinia

    DBLP profile ↗ORCID search ↗
20. Error Feedback Can Accurately Compress Preconditioners
    ICML 2024 ·

    Ionut-Vlad Modoranu

    DBLP profile ↗ORCID search ↗
21. Extreme Compression of Large Language Models via Additive Quantization
    ICML 2024 ·

    Vage Egiazarian

    DBLP profile ↗ORCID search ↗
22. L-GreCo: Layerwise-adaptive Gradient Compression For Efficient Data-parallel Deep Learning
    MLSys 2024 ·

    Ilia Markov

    DBLP profile ↗ORCID search ↗
23. Mathador-LM: A Dynamic Benchmark for Mathematical Reasoning on Large Language Models
    EMNLP 2024 ·

    Eldar Kurtic

    DBLP profile ↗ORCID search ↗
24. MicroAdam: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence
    NeurIPS 2024 ·

    Ionut-Vlad Modoranu

    DBLP profile ↗ORCID search ↗
25. PV-Tuning: Beyond Straight-Through Estimation for Extreme LLM Compression
    NeurIPS 2024 ·

    Vladimir Malinovskii

    DBLP profile ↗ORCID search ↗
26. QMoE: Sub-1-Bit Compression of Trillion Parameter Models
    MLSys 2024 ·

    Elias Frantar

    DBLP profile ↗ORCID search ↗
27. QUIK: Towards End-to-end 4-Bit Inference on Generative Large Language Models
    EMNLP 2024 ·

    Saleh Ashkboos

    DBLP profile ↗ORCID search ↗
28. QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs
    NeurIPS 2024 ·

    Saleh Ashkboos

    DBLP profile ↗ORCID search ↗
29. RoSA: Accurate Parameter-Efficient Fine-Tuning via Robust Adaptation
    ICML 2024 ·

    Mahdi Nikdan

    DBLP profile ↗ORCID search ↗
30. SPADE: Sparsity-Guided Debugging for Deep Neural Networks
    ICML 2024 ·

    Arshia Soltani Moakhar

    DBLP profile ↗ORCID search ↗
31. Scaling Laws for Sparsely-Connected Foundation Models
    ICLR 2024 ·

    Elias Frantar

    DBLP profile ↗ORCID search ↗
32. SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
    ICLR 2024 ·

    Tim Dettmers

    DBLP profile ↗ORCID search ↗
33. The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information
    NeurIPS 2024 ·

    Diyuan Wu

    DBLP profile ↗ORCID search ↗