Papers › Coevolving with the Other You: Fine-Tuning LLM with Sequential Cooperative Multi-Agent...

Coevolving with the Other You: Fine-Tuning LLM with Sequential Cooperative Multi-Agent Reinforcement Learning

8 Oct 2024arXiv:2410.06101archive 2025-07-28

Hao Ma, Tianyi Hu, Zhiqiang Pu, Boyin Liu, Xiaolin Ai, Yanyan Liang, Min Chen

Reinforcement learning (RL) has emerged as a pivotal technique for fine-tuning large language models (LLMs) on specific tasks. However, prevailing RL fine-tuning methods predominantly rely on PPO and its variants. Though these algorithms are effective in general RL settings, they often exhibit suboptimal performance and vulnerability to distribution collapse when applied to the fine-tuning of LLMs. In this paper, we propose CORY, extending the RL fine-tuning of LLMs to a sequential cooperative multi-agent reinforcement learning framework, to leverage the inherent coevolution and emergent capabilities of multi-agent systems. In CORY, the LLM to be fine-tuned is initially duplicated into two autonomous agents: a pioneer and an observer. The pioneer generates responses based on queries, while the observer generates responses using both the queries and the pioneer's responses. The two agents are trained together. During training, the agents exchange roles periodically, fostering cooperation and coevolution between them. Experiments evaluate CORY's performance by fine-tuning GPT-2 and Llama-2 under subjective and objective reward functions on the IMDB Review and GSM8K datasets, respectively. Results show that CORY outperforms PPO in terms of policy optimality, resistance to distribution collapse, and training robustness, thereby underscoring its potential as a superior methodology for refining LLMs in real-world applications.

PaperPDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2410.06101")

Code

Syntology Ran 6 of 11 code samples harvested from 1 repository linked to this paper; 5 have no recorded run. Of those that ran: 2 ran · our draft was wrong; 4 ran with no contract checked.

By repository: official repository: 11 samples from 1 repository, 6 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Harry67Hu/CORY officialmentioned on GitHubpytorchMIT 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

11 samples harvested; 6 ran; 0 honoured the contract we drafted; 5 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · our draft was wrong
4ran
5unverified

Licence: 0 of the 11 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 Harry67Hu/CORY. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

collator Harry67Hu/CORY/imdb_train/cory.py official repository ran · our draft was wrong MIT (permissive) · bda36ad0c44dbd74 · report
create_demo_text Harry67Hu/CORY/gsm8k_utils/gsm8k_utils.py official repository ran MIT (permissive) · 5013cfeaf23a03ad · report
extract_answer_from_output Harry67Hu/CORY/gsm8k_utils/gsm8k_utils.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · 96fe095fcc6b821a · report
is_correct Harry67Hu/CORY/gsm8k_utils/gsm8k_utils.py official repository ran MIT (permissive) · e198da7bdc5637e3 · report
respond_to_batch Harry67Hu/CORY/trl/models/model.py official repository ran MIT (permissive) · 3fe7529a76b17e9c · report
top_k_top_p_filtering Harry67Hu/CORY/trl/core.py official repository ran MIT (permissive) · 5bcf8e086423cf6b · report
build_dataset Harry67Hu/CORY/gsm8k_utils/gsm8k_eval.py official repository unverified MIT (permissive) · 0efa9af73116efc0 · report
convert_to_scalar Harry67Hu/CORY/trl/core.py official repository unverified MIT (permissive) · c32b636f13ed9ee5 · report
flatten_dict Harry67Hu/CORY/trl/core.py official repository unverified MIT (permissive) · a51ae45a1f6c0b5b · report
is_transformers_greater_than Harry67Hu/CORY/trl/import_utils.py official repository unverified MIT (permissive) · f4334934ee795269 · report
scheduler_step Harry67Hu/CORY/trl/models/modeling_sd_base.py official repository unverified MIT (permissive) · 59046fe10dd792f4 · report

Tasks

GSM8KMulti-agent Reinforcement LearningReinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdamAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDiscriminative Fine-TuningDropoutEntropy RegularizationGPT-2Layer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionPPOResidual ConnectionSoftmaxWeight Decay

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