When Quantization Affects Confidence of Large Language Models?

• URL: http://arxiv.org/abs/2405.00632v1
• Date: Wed, 1 May 2024 16:58:28 GMT
• Title: When Quantization Affects Confidence of Large Language Models?
• Authors: Irina Proskurina, Luc Brun, Guillaume Metzler, Julien Velcin,
• Abstract summary: We show that GPTQ to 4-bit results in a decrease in confidence regarding true labels, with varying impacts observed among different language models. We propose an explanation for quantization loss based on confidence levels, indicating that quantization disproportionately affects samples where the full model exhibited low confidence levels in the first place.
• Score: 4.338589334157708
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent studies introduced effective compression techniques for Large Language Models (LLMs) via post-training quantization or low-bit weight representation. Although quantized weights offer storage efficiency and allow for faster inference, existing works have indicated that quantization might compromise performance and exacerbate biases in LLMs. This study investigates the confidence and calibration of quantized models, considering factors such as language model type and scale as contributors to quantization loss. Firstly, we reveal that quantization with GPTQ to 4-bit results in a decrease in confidence regarding true labels, with varying impacts observed among different language models. Secondly, we observe fluctuations in the impact on confidence across different scales. Finally, we propose an explanation for quantization loss based on confidence levels, indicating that quantization disproportionately affects samples where the full model exhibited low confidence levels in the first place.

Related papers

• GWQ: Gradient-Aware Weight Quantization for Large Language Models [61.17678373122165]
  gradient-aware weight quantization (GWQ) is the first quantization approach for low-bit weight quantization that leverages gradients to localize outliers. GWQ retains the corresponding to the top 1% outliers preferentially at FP16 precision, while the remaining non-outlier weights are stored in a low-bit format. In the zero-shot task, GWQ quantized models have higher accuracy compared to other quantization methods.
  arXiv  Detail & Related papers  (2024-10-30T11:16:04Z)
• Scaling laws for post-training quantized large language models [41.78467383320145]
  Generalization abilities of well-trained large language models (LLMs) are known to scale predictably as a function of model size. The quality of LLMs after post-training compression remains highly unpredictable, often requiring case-by-case validation in practice.
  arXiv  Detail & Related papers  (2024-10-15T23:34:22Z)
• Advancing Multimodal Large Language Models with Quantization-Aware Scale Learning for Efficient Adaptation [70.22782550540714]
  Quantization-aware Scale LeArning method based on multimodal Warmup, termed QSLAW. We introduce a Quantization-aware Scale LeArning method based on multimodal Warmup, termed QSLAW.
  arXiv  Detail & Related papers  (2024-08-07T12:42:09Z)
• Cycles of Thought: Measuring LLM Confidence through Stable Explanations [53.15438489398938]
  Large language models (LLMs) can reach and even surpass human-level accuracy on a variety of benchmarks, but their overconfidence in incorrect responses is still a well-documented failure mode. We propose a framework for measuring an LLM's uncertainty with respect to the distribution of generated explanations for an answer.
  arXiv  Detail & Related papers  (2024-06-05T16:35:30Z)
• What Makes Quantization for Large Language Models Hard? An Empirical Study from the Lens of Perturbation [55.153595212571375]
  Quantization is a technique for improving the memory and computational efficiency of large language models (LLMs) We propose a new perspective on quantization, viewing it as perturbations added to the weights and activations of LLMs. We conduct experiments with various artificial perturbations to explore their impact on LLM performance.
  arXiv  Detail & Related papers  (2024-03-11T03:42:51Z)
• The Impact of Quantization on the Robustness of Transformer-based Text Classifiers [5.281054432963503]
  This work is the first application of quantization on the robustness of NLP models. We evaluate the impact of quantization on BERT and DistilBERT models in text classification using SST-2, Emotion, and MR datasets. Our experiments indicate that quantization increases the robustness of the model by 18.80% on average compared to adversarial training.
  arXiv  Detail & Related papers  (2024-03-08T14:55:05Z)
• Do Emergent Abilities Exist in Quantized Large Language Models: An Empirical Study [90.34226812493083]
  This work aims to investigate the impact of quantization on emphemergent abilities, which are important characteristics that distinguish LLMs from small language models. Our empirical experiments show that these emergent abilities still exist in 4-bit quantization models, while 2-bit models encounter severe performance degradation. To improve the performance of low-bit models, we conduct two special experiments: (1) fine-gained impact analysis that studies which components (or substructures) are more sensitive to quantization, and (2) performance compensation through model fine-tuning.
  arXiv  Detail & Related papers  (2023-07-16T15:11:01Z)
• PreQuant: A Task-agnostic Quantization Approach for Pre-trained Language Models [52.09865918265002]
  We propose a novel quantize before fine-tuning'' framework, PreQuant. PreQuant is compatible with various quantization strategies, with outlier-aware fine-tuning incorporated to correct the induced quantization error. We demonstrate the effectiveness of PreQuant on the GLUE benchmark using BERT, RoBERTa, and T5.
  arXiv  Detail & Related papers  (2023-05-30T08:41:33Z)
• Mixed-Precision Inference Quantization: Radically Towards Faster inference speed, Lower Storage requirement, and Lower Loss [4.877532217193618]
  Existing quantization techniques rely heavily on experience and "fine-tuning" skills. This study provides a methodology for acquiring a mixed-precise quantization model with a lower loss than the full precision model. In particular, we will demonstrate that neural networks with massive identity mappings are resistant to the quantization method.
  arXiv  Detail & Related papers  (2022-07-20T10:55:34Z)
• An Investigation on Different Underlying Quantization Schemes for Pre-trained Language Models [33.49417100179159]
  We implement k-means quantization and compare its performance on the fix-precision quantization of BERT with linear quantization. We also compare the two quantization schemes on ALBERT models to explore the robustness differences between different pre-trained models.
  arXiv  Detail & Related papers  (2020-10-14T14:05:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.