SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.01543 · CVPR · 2024

Retraining-free Model Quantization via One-Shot Weight-Coupling Learning

Wenwu Zhu, Chen Tang, Xinzhu Ma, Zhi Wang, Yuan Meng, Jiacheng Jiang, Rongwei Lu, Shuzhao Xie

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2401.01543.json · badge: /api/ran/2401.01543.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
1hunters/retraining-free-quantization community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
QuanConv2d Ran
ran on a synthesized input · constructed an object; no result was computed
1hunters/retraining-free-quantization/quan/func.py
code served (permissive licence) · get_code("8534467a1f00fe8d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, QuanConv2d: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple quantization function class
class SimpleQuantFn:
    def __init__(self):
        self.scale = None
        self.zero_point = None
    
    def init_from(self, data, bits_list):
        # Initialize quantization parameters
        if data is not None:
            self.scale = data.abs().max() / 127.0
        else:
            self.scale = 1.0
        self.zero_point = 0
    
    def __call__(self, x, bits, is_activation=False):
        # Simple quantization: just scale and clip
        if bits == 32:
            return x
        max_val = (2 ** (bits - 1)) - 1
        return torch.clamp(x / self.scale, -max_val, max_val) * self.scale
    
    def sample(self, bit_cands, max=False, min=False):
        # Sample a bit width from candidates
        bit_cands_list = bit_cands.tolist() if isinstance(bit_cands, torch.Tensor) else list(bit_cands)
        if max:
            return max(bit_cands_list)
        if min:
            return min(bit_cands_list)
        return bit_cands_list[0]

# Create a base Conv2d layer to wrap
base_conv = nn.Conv2d(in_channels=3, out_channels=4, kernel_size=3, stride=1, padding=1, bias=True)

# Create quantization function instances
quan_w_fn = SimpleQuantFn()
quan_a_fn = SimpleQuantFn()

# Define all arguments for QuanConv2d
ARGS = {
    'm': base_conv,
    'bits_list': [8],
    'quan_w_fn': quan_w_fn,
    'quan_a_fn': quan_a_fn,
    'fixed_bits': None,
    'split_aw_cands': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type QuanConv2d — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a QuanConv2d output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Quantization is of significance for compressing the overparameterized deep neural models and deploying them on resource-limited devices. Fixed-precision quantization suffers from performance drop due to the limited numerical representation ability. Conversely, mixed-precision quantization (MPQ) is advocated to compress the model effectively by allocating heterogeneous bit-width for layers. MPQ is typically organized into a searching-retraining twostage process. Previous works only focus on determining the optimal bit-width configuration in the first stage efficiently, while ignoring the considerable time costs in the second stage and thus hindering deployment efficiency significantly. In this paper, we devise a one-shot trainingsearching paradigm for mixed-precision model compression. Specifically, in the first stage, all potential bit-width configurations are coupled and thus optimized simultaneously within a set of shared weights. However, our observations reveal a previously unseen and severe bit-width interference phenomenon among highly coupled weights during optimization, leading to considerable performance degradation under a high compression ratio. To tackle this problem, we first design a bit-width scheduler to dynamically freeze the most turbulent bit-width of layers during training, to ensure the rest bit-widths converged properly. Then, taking inspiration from information theory, we present an information distortion mitigation technique to align the behaviour of the bad-performing bit-widths to the well-performing ones. In the second stage, an inferenceonly greedy search scheme is devised to evaluate the goodness of configurations without introducing any additional training costs. Extensive experiments on three representative models and three datasets demonstrate the effectiveness of the proposed method. Code can be available on https://github.com/1hunters/retraining-free-quantization.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2401.01543")
get_code_for_paper("2401.01543")
have("2401.01543")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2401.01543.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2401.01543.svg)](https://syntology.ai/paper/2401.01543)

Connect an agent — have() is free.