Pith. sign in

REVIEW 10 cited by

Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2505.01043 v2 pith:LQRU3AGD submitted 2025-05-02 cs.LG cs.AI

Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities

classification cs.LG cs.AI
keywords traininglow-precisionmethodsefficiencyapproachesfieldformatshardware
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Large language models (LLMs) have achieved impressive performance across various domains. However, the substantial hardware resources required for their training present a significant barrier to efficiency and scalability. To mitigate this challenge, low-precision training techniques have been widely adopted, leading to notable advancements in training efficiency. Despite these gains, low-precision training involves several components, such as weights, activations, and gradients, each of which can be represented in different numerical formats. The resulting diversity has created a fragmented landscape in low-precision training research, making it difficult for researchers to gain a unified overview of the field. This survey provides a comprehensive review of existing low-precision training methods. To systematically organize these approaches, we categorize them into three primary groups based on their underlying numerical formats, which is a key factor influencing hardware compatibility, computational efficiency, and ease of reference for readers. The categories are (1) fixed-point and integer-based methods, (2) floating-point-based methods, and (3) customized format-based methods. Additionally, we discuss quantization-aware training approaches, which share key similarities with low-precision training during forward propagation. Beyond efficiency, we examine robustness and deployment reliability under low precision. Finally, we highlight several promising research directions to advance this field. A collection of papers discussed in this survey is provided in https://github.com/Hao840/Awesome-Low-Precision-Training.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

    cs.LG 2025-10 unverdicted novelty 7.0

    Low-precision Flash Attention fails due to similar low-rank attention representations combined with biased rounding errors that accumulate and corrupt weight updates; a minimal fix to reduce rounding bias stabilizes training.

  2. Stable FP4 Training via Transposition-Invariant Block Quantization

    cs.LG 2026-07 conditional novelty 6.0

    Transposition-invariant 2D block FP4 quantization plus truncation-free scaling and stochastic rounding enables stable end-to-end FP4 LLM training within ~1% of BF16.

  3. LC-QAT: Data-Efficient 2-Bit QAT for LLMs via Linear-Constrained Vector Quantization

    cs.CL 2026-06 unverdicted novelty 6.0

    LC-QAT achieves data-efficient 2-bit weight-only QAT for LLMs by representing quantized weights as a learned affine transform over discrete vectors, supporting end-to-end optimization from a high-quality PTQ start.

  4. Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor

    cs.LG 2026-05 unverdicted novelty 6.0

    MXFP4 quantization error decomposes into scale bias, deadzone truncation, and grid noise; mode-targeted corrections recover BF16 accuracy within 0.7% on Qwen2.5-3B and exceed it by 1.0% on Qwen3-30B-A3B.

  5. Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor

    cs.LG 2026-05 unverdicted novelty 6.0

    MXFP4 quantization error decomposes into scale bias, deadzone truncation, and grid noise; macro-block scaling, outlier fallback, and adaptive quantization noise recover BF16 accuracy to within 0.7% and 3.0% on tested models.

  6. Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, recoverable deadzone, and an irreducible floor

    cs.LG 2026-05 unverdicted novelty 6.0

    MXFP4 error decomposes into scale bias, deadzone truncation, and grid noise that each dominate distinct RL failure modes, with macro-block scaling, outlier fallback, and adaptive noise recovering or exceeding BF16 per...

  7. StoSignSGD: Unbiased Structural Stochasticity Fixes SignSGD for Training Large Language Models

    cs.LG 2026-04 unverdicted novelty 6.0

    StoSignSGD resolves SignSGD divergence on non-smooth objectives via structural stochasticity, matching optimal convex rates and improving non-convex bounds while delivering 1.44-2.14x speedups in FP8 LLM pretraining.

  8. Reliable Evaluation Protocol for Low-Precision Retrieval

    cs.IR 2025-08 unverdicted novelty 6.0

    Proposes High-Precision Scoring (HPS) and Tie-aware Retrieval Metrics (TRM) to reduce tie-induced instability in low-precision retrieval evaluation.

  9. LC-QAT: Data-Efficient 2-Bit QAT for LLMs via Linear-Constrained Vector Quantization

    cs.CL 2026-06 unverdicted novelty 5.0

    LC-QAT is a 2-bit weight-only vector quantization aware training framework for LLMs that uses linear-constrained affine mappings to achieve data-efficient optimization and outperform prior QAT methods.

  10. PowLU: An Activation Function for Stable Pre-Training of LLMs

    cs.CL 2026-05 unverdicted novelty 4.0

    PowLU replaces SwiGLU with a rational-power activation to reduce outlier amplification and numerical instability during large-scale LLM pre-training while matching performance.