mlsanityData to 14 Oct 2026

Robust Ultra Low-Bit Post-Training Quantization via Stable Diagonal Curvature Estimate

arXiv 2604.13806 ↗First posted 15 Apr 2026cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Large Language Models (LLMs) are widely used across many domains, but their scale makes deployment challenging. Post-Training Quantization (PTQ) reduces memory footprint without retraining by leveraging a small calibration set. Recent Hessian-based PTQ methods compensate quantization error via cross-channel dependencies, but such approaches degrade at low bit-widths due to noisy curvature estimates from limited calibration data. We propose DASH-Q, a robust PTQ framework using diagonal Hessian approximation and iterative weighted least squares. By discarding noise-prone dependencies, DASH-Q filters sampling noise while prioritizing the preservation of salient feature power. We outperform other PTQ baselines in ultra low-bit regime, improving zero-shot accuracy by 7.01% on average and up to 14.01% over the strongest baselines across five baseline LLM models, while showing robust and stable performance with very small calibration data.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 39 of its 50 references that are in the map. First posted on arXiv 15 Apr 2026.

Builds on

39 of 50 references.

References · 39

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2025Qwen3 Technical Reportcited by 8,820
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  5. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,886
  6. 2019HellaSwag: Can a Machine Really Finish Your Sentence?cited by 4,100
  7. 2016Pointer Sentinel Mixture Modelscited by 3,096
  8. 2019PIQA: Reasoning about Physical Commonsense in Natural Languagecited by 2,883
  9. 2019BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questionscited by 2,183
  10. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,164
  11. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  12. 2024Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phonecited by 1,917
27 more
  1. 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,868
  2. 2024Mixtral of Expertscited by 1,823
  3. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  4. 2022SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Modelscited by 1,386
  5. 2019ZeRO: Memory Optimizations Toward Training Trillion Parameter Modelscited by 1,301
  6. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,298
  7. 2024DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Modelscited by 867
  8. 2024QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMscited by 634
  9. 2023QuIP: 2-Bit Quantization of Large Language Models With Guaranteescited by 414
  10. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  11. 2022Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruningcited by 322
  12. 2024The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bitscited by 308
  13. 2023Atom: Low-bit Quantization for Efficient and Accurate LLM Servingcited by 272
  14. 2019WinoGrande: An Adversarial Winograd Schema Challenge at Scalecited by 204
  15. 2023RPTQ: Reorder-based Post-training Quantization for Large Language Modelscited by 116
  16. 2023OWQ: Outlier-Aware Weight Quantization for Efficient Fine-Tuning and Inference of Large Language Modelscited by 114
  17. 2014A Stochastic Quasi-Newton Method for Large-Scale Optimizationcited by 69
  18. 2023On the Impact of Calibration Data in Post-training Quantization and Pruningcited by 58
  19. 2024MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Modelscited by 52
  20. 2024ExeGPT: Constraint-Aware Resource Scheduling for LLM Inferencecited by 45
  21. 2025Quantization Error Propagation: Revisiting Layer-Wise Post-Training Quantizationcited by 41
  22. 2025Bridging the Gap Between Promise and Performance for Microscaling FP4 Quantizationcited by 32
  23. 2021On the Variance of the Fisher Information for Deep Learningcited by 22
  24. 2024Exploring the Trade-Offs: Quantization Methods, Task Difficulty, and Model Size in Large Language Models From Edge to Giantcited by 14
  25. 2024Trade-Offs of Diagonal Fisher Information Matrix Estimatorscited by 10
  26. 2025FlexiQ: Adaptive Mixed-Precision Quantization for Latency/Accuracy Trade-Offs in Deep Neural Networkscited by 1
  27. 2026HeRo-Q: A General Framework for Stable Low Bit Quantization via Hessian Conditioningcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Beyond Expert Rankings: Contrast Calibration for Mixed-Precision MoE Quantizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025PTQ1.61: Push the Real Limit of Extremely Low-Bit Post-Training Quantization Methods for Large Language Models
  2. 2024AdpQ: A Zero-shot Calibration Free Adaptive Post Training Quantization Method for LLMs
  3. 2026G2PTQ: Improving LLM Post-Training Quantization with Generalized Gradient Compensation
  4. 2025Rethinking Post-Training Quantization: Introducing a Statistical Pre-Calibration Approach
  5. 2026All for 1-Bit: Towards Genuine 1-Bit Post-Training Quantization for LLMs
  6. 2026C-PTQ: Fisher-weighted Channel-wise Sensitivity for Post-training Quantization of MLLMs
  7. 2023ZeroQuant-V2: Exploring Post-training Quantization in LLMs from Comprehensive Study to Low Rank Compensation
  8. 2026Recurrent Residual Quantization: A Progressive Multi-Precision Representation for LLMs