Papers › Weighted-Reward Preference Optimization for Implicit Model Fusion

Weighted-Reward Preference Optimization for Implicit Model Fusion

4 Dec 2024arXiv:2412.03187archive 2025-07-28

ZiYi Yang, Fanqi Wan, Longguang Zhong, Tianyuan Shi, Xiaojun Quan

While fusing heterogeneous open-source LLMs with varying architectures and sizes can potentially integrate the strengths of different models, existing fusion methods face significant challenges, such as vocabulary alignment and merging distribution matrices. These procedures are not only complex but also prone to introducing noise and errors. In this paper, we propose an implicit fusion method, Weighted-Reward Preference Optimization (WRPO), which leverages preference optimization between the source LLMs and the target LLM to transfer their capabilities effectively. WRPO eliminates the need for vocabulary alignment and matrix fusion and can be efficiently scaled to accommodate various LLMs. To address distributional deviations between the source and target LLMs, WRPO introduces a progressive adaptation strategy that gradually shifts reliance on preferred examples from the target LLM to the source LLMs. Extensive experiments on the MT-Bench, AlpacaEval-2, and Arena-Hard benchmarks demonstrate that WRPO consistently outperforms existing knowledge fusion methods and various fine-tuning baselines. When applied to LLaMA3-8B-Instruct as the target model, WRPO achieves a length-controlled win rate of 55.9% against GPT-4-Preview-1106 on AlpacaEval-2 and a win rate of 46.2% against GPT-4-0314 on Arena-Hard. Our code is available at https://github.com/SLIT-AI/WRPO.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 14 code samples harvested from 3 repositories linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 4 violated, 0 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 1 sample from 1 repository, 1 ran (0 with no instrument failure; 1 where Syntology's instrument failed); community (archive-listed): 13 samples from 2 repositories, 8 ran (4 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 9 samples that ran has no recorded date.

SLIT-AI/WRPO officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
fanqiwan/fuseai mentioned on GitHubpytorchSyntology: 10 samples, 7 ran (3 instrument-failed) · 3 unverified report
fanqiwan/fusellm mentioned on GitHubpytorchSyntology: 3 samples, 1 ran (1 instrument-failed) · 2 unverified report
slit-ai/fusechat-3.0 mentioned on GitHubSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

14 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 4 violated, 0 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

4ran · violated contract
5ran · our draft was wrong
5unverified

Licence: 14 of the 14 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

SLIT-AI/WRPO official repository 1 sample, 1 ran (1 instrument-failed) · 0 unverified
pad_to_length in 18 papers scripts/wrpo_trainer.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 030b2d5e6e147a1d · report fixture: signature tensor:Tensor/2/float, length:int/0/int, pad_value:float/0/float, dim:int/0/int · returned Tensor · level recorded, date not recorded
fanqiwan/fuseai community (archive-listed) 10 samples, 7 ran (3 instrument-failed) · 3 unverified
check_nan in 6 papers FuseChat/train/filter_nan.py ran · violated contract no licence file found · pointer only · 524b27f03377a248 · report fixture: signature example:dict/0/? · returned bool · level recorded 2026-09-24
check_nan_base FuseChat/train/filter_nan.py ran · violated contract no licence file found · pointer only · fd32354386c6ce83 · report fixture: signature example:dict/0/? · returned bool · level recorded 2026-09-24
dict_to_list in 8 papers FuseChat/train/align_token_and_vocab.py ran · our draft was wrong no licence file found · pointer only · 98b9e3b2ff2e30b8 · report fixture: signature examples:dict/0/? · returned list · level recorded 2026-09-24
generate_completions in 38 papers FuseO1-Preview/math_evaluation/model_utils.py ran · our draft was wrong no licence file found · pointer only · 1525d8bc0fc8d5a0 · report fixture: signature model:callable/0/?, tokenizer:callable/0/?, prompts:list/1/str, batch_size:int/0/int, stop_id_sequences:list/1/str, add_special_tokens:bool/0/bool, disable_tqdm:bool/0/bool, max_length:int/0/int, num_return_sequences:int/0/int · returned list · level recorded 2026-09-24
list_to_dict in 8 papers FuseChat/train/align_token_and_vocab.py ran · our draft was wrong no licence file found · pointer only · 8e08bca34751633c · report fixture: signature examples:list/1/dict · returned dict · level recorded 2026-09-24
raise_warning_for_incompatible_cpu_offloading_configuration in 16 papers FuseChat/train/model/model_adapter.py ran · violated contract no licence file found · pointer only · 0bf414a2b44babe6 · report fixture: signature device:str/0/str, load_8bit:bool/0/bool, cpu_offloading:bool/0/bool · returned bool · level recorded 2026-09-22
sigmoid in 216 papers
census: 36 implementations from 56 papers, 7 distinct outputs across 3 buckets, one shared input per bucket
FuseChat/train/align_token_and_vocab.py ran · violated contract fingerprinted no licence file found · pointer only · d0b132e4577e5bbb · report fixture: signature x:ndarray/1/float · returned ndarray · level recorded 2026-09-24
get_conv_template in 72 papers FuseChat/train/conversation.py unverified no licence file found · pointer only · 502fd3534e0509c1 · report
get_gpu_memory in 16 papers FuseChat/train/model/model_adapter.py unverified no licence file found · pointer only · 4bec2455307c11b7 · report
load_hf_lm_and_tokenizer in 21 papers FuseO1-Preview/math_evaluation/model_utils.py unverified no licence file found · pointer only · ee71a94d9dd1b8e9 · report
fanqiwan/fusellm community (archive-listed) 3 samples, 1 ran (1 instrument-failed) · 2 unverified
get_logger in 383 papers FuseLLM/src/utils/others.py ran · our draft was wrong no licence file found · pointer only · 1082508e65f77331 · report fixture: signature name:str/0/str · returned Logger · level recorded 2026-09-24
get_base_model in 10 papers FuseLLM/src/utils/others.py unverified no licence file found · pointer only · df472be23644090f · report
get_tokenizer in 156 papers FuseLLM/src/utils/others.py unverified no licence file found · pointer only · 503f70e78560aa9e · report

Tasks

model

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections