SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.04070 · ICLR · 2025

Personalized Alignment of LLMs at Decoding-Time PAD: PERSONALIZED ALIGNMENT OF LLMS AT DECODING-TIME

Zuozhu Liu, Ruizhe Chen, Wenhao Chai, Xiaotian Zhang, Meng Luo

arXiv · PDF · Open in the Atlas

Code that ran

4 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 2 failures of our instrument; 4 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/2410.04070.json · badge: /api/ran/2410.04070.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
YangRui2015/RiC canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
srzer/MOD canonical (classifier label)
classifier basis: census
1 ran · 1 no recorded run
FunctionStatusWhere it lives
convert_binary_pil Ran
ran, honoured the contract we drafted
YangRui2015/RiC/ric_txt2img/image_metric.py
pointer only (licence: NONE) · get_code("28d5e50dbbf83280")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YangRui2015/RiC when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

encode_jpeg Ran
ran; our contract draft was wrong, not the code
YangRui2015/RiC/ric_txt2img/image_metric.py
pointer only (licence: NONE) · get_code("5dc2d5b477d36a34")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YangRui2015/RiC when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_pareto_points Ran
ran, violated the contract we drafted
YangRui2015/RiC/ric_txt2img/visualize_pareto.py
pointer only (licence: NONE) · get_code("939f2e01fbf8d40d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YangRui2015/RiC when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

stabilized_log1pexp Ran
ran; our contract draft was wrong, not the code
srzer/MOD/experiment-DPO/src/trainer/dpo_trainer.py
code served (permissive licence) · get_code("5100040c5aa125b4")
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 Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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

torch.manual_seed(0)

# The function takes a single parameter 'x' which is a tensor
# It performs element-wise operations on the tensor
# We need to provide a tensor with various values to test the numerical stability

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
build_dataset_with_preference Not yet run
unverified (harvested, no recorded run)
YangRui2015/RiC/ric/utils.py
pointer only (licence: NONE) · get_code("406b59a243695a5f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YangRui2015/RiC when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

build_dataset_with_preference_n Not yet run
unverified (harvested, no recorded run)
YangRui2015/RiC/ric/utils.py
pointer only (licence: NONE) · get_code("c2e52a32dd1f6431")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YangRui2015/RiC when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

build_summary_dataset_with_preference_n Not yet run
unverified (harvested, no recorded run)
YangRui2015/RiC/ric/utils.py
pointer only (licence: NONE) · get_code("384dbf3efb62957c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YangRui2015/RiC when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

relabel_dataset Not yet run
unverified (harvested, no recorded run)
srzer/MOD/experiment-DPO/src/utils/util_dataset.py
code served (permissive licence) · get_code("4981195135c64c27")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 8 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 2 later) and 0 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.

FunctionAlso present inEarliest-datedOthers named
build_dataset_with_preference 1 other paper
1 dated earlier
2402.10207 (2024-02)
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
build_dataset_with_preference_n 1 other paper
1 dated earlier
2402.10207 (2024-02)
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
build_summary_dataset_with_preference_n 1 other paper
1 dated earlier
2402.10207 (2024-02)
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
convert_binary_pil 1 other paper
1 dated earlier
2402.10207 (2024-02)
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
encode_jpeg 1 other paper
1 dated earlier
2402.10207 (2024-02)
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
find_pareto_points 1 other paper
1 dated earlier
2402.10207 (2024-02)
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
relabel_dataset 1 other paper
1 later
2505.19770 (2025-05)
Understanding the Performance Gap in Preference Learning: A Dichotomy of RLHF and DPO
stabilized_log1pexp 1 other paper
1 later
2505.19770 (2025-05)
Understanding the Performance Gap in Preference Learning: A Dichotomy of RLHF and DPO

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

Aligning with personalized preferences, which vary significantly across cultural, educational, and political differences, poses a significant challenge due to the computational costs and data demands of traditional alignment methods. In response, this paper presents Personalized Alignment at Decoding-time (PAD), a novel framework designed to align LLM outputs with diverse personalized preferences during the inference phase, eliminating the need for additional training. By introducing a unique personalized reward modeling strategy, this framework decouples the text generation process from personalized preferences, facilitating the generation of generalizable token-level personalized rewards. The PAD algorithm leverages these rewards to guide the decoding process, dynamically tailoring the base model's predictions to personalized preferences. Extensive experimental results demonstrate that PAD not only outperforms existing training-based alignment methods in terms of aligning with diverse preferences but also shows significant generalizability to preferences unseen during training and scalability across different base models. This work advances the capability of LLMs to meet user needs in real-time applications, presenting a substantial step forward in personalized LLM alignment. Our model and code are available here. Recent advancements have demonstrated success in aligning language models with human preferences and values (

For agents

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

get_harvested_code_for_paper("2410.04070")
get_code_for_paper("2410.04070")
have("2410.04070")

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

curl https://syntology.ai/api/ran/2410.04070.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.04070.svg)](https://syntology.ai/paper/2410.04070)

Connect an agent — have() is free.