OTARo: Once Tuning for All Precisions toward Robust On-Device LLMs

• URL: http://arxiv.org/abs/2511.13147v1
• Date: Mon, 17 Nov 2025 08:56:27 GMT
• Title: OTARo: Once Tuning for All Precisions toward Robust On-Device LLMs
• Authors: Shaoyuan Chen, Zhixuan Chen, Dawei Yang, Zhihang Yuan, Qiang Wu,
• Abstract summary: OTARo is a novel method that enables on-device Large Language Models to flexibly switch quantization precisions.<n>It achieves consistently strong and robust performance for all precisions.
• Score: 21.55040910903597
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) fine-tuning techniques not only improve the adaptability to diverse downstream tasks, but also mitigate adverse effects of model quantization. Despite this, conventional quantization suffers from its structural limitation that hinders flexibility during the fine-tuning and deployment stages. Practical on-device tasks demand different quantization precisions (i.e. different bit-widths), e.g., understanding tasks tend to exhibit higher tolerance to reduced precision compared to generation tasks. Conventional quantization, typically relying on scaling factors that are incompatible across bit-widths, fails to support the on-device switching of precisions when confronted with complex real-world scenarios. To overcome the dilemma, we propose OTARo, a novel method that enables on-device LLMs to flexibly switch quantization precisions while maintaining performance robustness through once fine-tuning. OTARo introduces Shared Exponent Floating Point (SEFP), a distinct quantization mechanism, to produce different bit-widths through simple mantissa truncations of a single model. Moreover, to achieve bit-width robustness in downstream applications, OTARo performs a learning process toward losses induced by different bit-widths. The method involves two critical strategies: (1) Exploitation-Exploration Bit-Width Path Search (BPS), which iteratively updates the search path via a designed scoring mechanism; (2) Low-Precision Asynchronous Accumulation (LAA), which performs asynchronous gradient accumulations and delayed updates under low bit-widths. Experiments on popular LLMs, e.g., LLaMA3.2-1B, LLaMA3-8B, demonstrate that OTARo achieves consistently strong and robust performance for all precisions.

Related papers

• QTALE: Quantization-Robust Token-Adaptive Layer Execution for LLMs [0.0]
  Large language models (LLMs) demand substantial computational and memory resources.<n>We propose QTALE, a novel framework that enables seamless integration of token-adaptive execution with quantization.
  arXiv  Detail & Related papers  (2026-02-11T02:19:11Z)
• RaBiT: Residual-Aware Binarization Training for Accurate and Efficient LLMs [5.782015253162346]
  Residual binarization enables matmul-free inference by stacking binary layers.<n>We propose RaBiT, a novel quantization framework that resolves coadaptation by algorithmically enforcing a residual hierarchy.<n>RaBiT achieves state-of-the-art performance, rivals even hardware-intensive Vector Quantization (VQ) methods, and delivers a $4.49times$ inference speed-up over full-precision models.
  arXiv  Detail & Related papers  (2026-02-05T06:41:11Z)
• AnyBCQ: Hardware Efficient Flexible Binary-Coded Quantization for Multi-Precision LLMs [14.922926621722235]
  We present AnyBCQ, a hardware-friendly multi-precision extension of Binary-Coded Quantization (BCQ)<n>Our progressive precision expansion mechanism incrementally refines scaling factors while reusing previously assigned binary codes.<n>Experiments demonstrate that AnyBCQ significantly narrows the accuracy drop in the low-bit regime.
  arXiv  Detail & Related papers  (2025-10-12T06:20:38Z)
• PT$^2$-LLM: Post-Training Ternarization for Large Language Models [52.4629647715623]
  Large Language Models (LLMs) have shown impressive capabilities across diverse tasks, but their large memory and compute demands hinder deployment.<n>We propose PT$2$-LLM, a post-training ternarization framework tailored for LLMs.<n>At its core is an Asymmetric Ternary Quantizer equipped with a two-stage refinement pipeline.
  arXiv  Detail & Related papers  (2025-09-27T03:01:48Z)
• Progressive Element-wise Gradient Estimation for Neural Network Quantization [2.1413624861650358]
  Quantization-Aware Training (QAT) methods rely on the Straight-Through Estimator (STE) to address the non-differentiability of discretization functions.<n>We propose Progressive Element-wise Gradient Estimation (PEGE) to address discretization errors between continuous and quantized values.<n>PEGE consistently outperforms existing backpropagation methods and enables low-precision models to match or even outperform the accuracy of their full-precision counterparts.
  arXiv  Detail & Related papers  (2025-08-27T15:59:36Z)
• MPQ-DMv2: Flexible Residual Mixed Precision Quantization for Low-Bit Diffusion Models with Temporal Distillation [74.34220141721231]
  We present MPQ-DMv2, an improved textbfMixed textbfPrecision textbfQuantization framework for extremely low-bit textbfDiffusion textbfModels.
  arXiv  Detail & Related papers  (2025-07-06T08:16:50Z)
• RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models [53.571195477043496]
  We propose an algorithm named Rotated Straight-Through-Estimator (RoSTE)<n>RoSTE combines quantization-aware supervised fine-tuning (QA-SFT) with an adaptive rotation strategy to reduce activation outliers.<n>Our findings reveal that the prediction error is directly proportional to the quantization error of the converged weights, which can be effectively managed through an optimized rotation configuration.
  arXiv  Detail & Related papers  (2025-02-13T06:44:33Z)
• SKIM: Any-bit Quantization Pushing The Limits of Post-Training Quantization [7.198819240352308]
  Large Language Models (LLMs) exhibit impressive performance across various tasks, but deploying them for inference poses challenges.<n>We propose a new method called SKIM: Scaled K-means clustering wIth Mixed precision.<n>In terms of model perplexity, our method narrows the gap between 3-bit quantized LLaMA models and their full precision counterparts by 16.3% on average.
  arXiv  Detail & Related papers  (2024-12-05T14:19:59Z)
• Progressive Mixed-Precision Decoding for Efficient LLM Inference [49.05448842542558]
  We introduce Progressive Mixed-Precision Decoding (PMPD) to address the memory-boundedness of decoding.<n>PMPD achieves 1.4$-$12.2$times$ speedup in matrix-vector multiplications over fp16 models.<n>Our approach delivers a throughput gain of 3.8$-$8.0$times$ over fp16 models and up to 1.54$times$ over uniform quantization approaches.
  arXiv  Detail & Related papers  (2024-10-17T11:46:33Z)
• DB-LLM: Accurate Dual-Binarization for Efficient LLMs [83.70686728471547]
  Large language models (LLMs) have significantly advanced the field of natural language processing. Existing ultra-low-bit quantization always causes severe accuracy drops. We propose a novel Dual-Binarization method for LLMs, namely DB-LLM.
  arXiv  Detail & Related papers  (2024-02-19T09:04:30Z)
• CBQ: Cross-Block Quantization for Large Language Models [66.82132832702895]
  Post-training quantization (PTQ) has played a key role in compressing large language models (LLMs) with ultra-low costs.<n>We propose CBQ, a cross-block reconstruction-based PTQ method for LLMs.<n> CBQ employs a cross-block dependency using a reconstruction scheme, establishing long-range dependencies across multiple blocks to minimize error accumulation.
  arXiv  Detail & Related papers  (2023-12-13T07:56:27Z)
• PIPE : Parallelized Inference Through Post-Training Quantization Ensembling of Residual Expansions [23.1120983784623]
  PIPE is a quantization method that leverages residual error expansion, along with group sparsity and an ensemble approximation for better parallelization. It achieves superior performance on every benchmarked application (from vision to NLP tasks), architecture (ConvNets, transformers) and bit-width.
  arXiv  Detail & Related papers  (2023-11-27T13:29:34Z)
• SqueezeLLM: Dense-and-Sparse Quantization [80.32162537942138]
  Main bottleneck for generative inference with LLMs is memory bandwidth, rather than compute, for single batch inference. We introduce SqueezeLLM, a post-training quantization framework that enables lossless compression to ultra-low precisions of up to 3-bit. Our framework incorporates two novel ideas: (i) sensitivity-based non-uniform quantization, which searches for the optimal bit precision assignment based on second-order information; and (ii) the Dense-and-Sparse decomposition that stores outliers and sensitive weight values in an efficient sparse format.
  arXiv  Detail & Related papers  (2023-06-13T08:57:54Z)

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.