SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.09907 · ICML · 2026

RADAR: Redundancy-Aware Diffusion for Multi-Agent Communication Structure Generation

Wei Ji, Juncheng Li, Zhen Zhang, Hao Fei, Jun Wen, Wanjing Zhou

arXiv · PDF · Open in the Atlas

Code that ran

5 of 10 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; 1 honoured the contract we drafted, 1 violated it and 1 has no contract verdict; 2 failures of our instrument; 5 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/2605.09907.json · badge: /api/ran/2605.09907.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
bingreeky/MaAS canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
yanweiyue/AgentPrune canonical (classifier label)
classifier basis: census
1 ran · 0 no recorded run
yanweiyue/GDesigner canonical (classifier label)
classifier basis: census
1 ran · 0 no recorded run
tsinghua-fib-lab/AgentSquare canonical (classifier label)
classifier basis: census
0 ran · 4 no recorded run
cszhangzhen/RADAR community
Papers with Code archive
1 ran · 1 no recorded run
FunctionStatusWhere it lives
NodeMasking Ran
ran on a synthesized input · constructed an object; no result was computed
cszhangzhen/RADAR/model/gd.py
pointer only (licence: NONE) · get_code("6240d7aad9cb1553")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cszhangzhen/RADAR 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_strings_between_pluses Ran
ran; our contract draft was wrong, not the code
yanweiyue/AgentPrune/AgentPrune/agents/analyze_agent.py
pointer only (licence: NONE); the stored copy get_code serves, from yao-guan/MOC, is pointer only (licence: NONE) · get_code("e785d76257026524")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yanweiyue/AgentPrune 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.

is_func Ran
ran, violated the contract we drafted
bingreeky/MaAS/maas/repo_parser.py
pointer only (licence: NONE) · get_code("84e834cc9e6e17da")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bingreeky/MaAS 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.

merge_dict Ran
ran; our contract draft was wrong, not the code
bingreeky/MaAS/maas/config2.py
pointer only (licence: NONE) · get_code("b2eab098d73493e2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, bingreeky/MaAS 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.

min_max_norm Ran
ran, honoured the contract we drafted
yanweiyue/GDesigner/GDesigner/graph/graph.py
pointer only (licence: NONE); the stored copy get_code serves, from yao-guan/MOC, is pointer only (licence: NONE) · get_code("b4a55c20a81f5527")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, yanweiyue/GDesigner 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.

GDFramework Not yet run
unverified (harvested, no recorded run)
cszhangzhen/RADAR/model/gd.py
pointer only (licence: NONE) · get_code("a8bf979767b75335")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cszhangzhen/RADAR 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.

get_prompt_memory Not yet run
unverified (harvested, no recorded run)
tsinghua-fib-lab/AgentSquare/module_evolution/prompt_memory.py
pointer only (licence: NONE) · get_code("b6ee8ce4f463d87e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, tsinghua-fib-lab/AgentSquare 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.

get_prompt_planning Not yet run
unverified (harvested, no recorded run)
tsinghua-fib-lab/AgentSquare/module_evolution/prompt_planning.py
pointer only (licence: NONE) · get_code("536bd0213afd6b66")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, tsinghua-fib-lab/AgentSquare 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.

get_prompt_reasoning Not yet run
unverified (harvested, no recorded run)
tsinghua-fib-lab/AgentSquare/module_evolution/prompt_reasoning.py
pointer only (licence: NONE) · get_code("72a558fbb675cc91")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, tsinghua-fib-lab/AgentSquare 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.

get_prompt_tooluse Not yet run
unverified (harvested, no recorded run)
tsinghua-fib-lab/AgentSquare/module_evolution/prompt_tooluse.py
pointer only (licence: NONE) · get_code("e56531059f4368b2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, tsinghua-fib-lab/AgentSquare 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.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 2 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
find_strings_between_pluses 1 other paper
1 later
2606.02359 (2026-06)
MOC: Multi-Order Communication in LLM-based Multi-Agent Systems
get_prompt_memory 1 other paper
1 dated earlier
2410.06153 (2024-10)
AgentSquare: Automatic LLM Agent Search in Modular Design Space
get_prompt_planning 1 other paper
1 dated earlier
2410.06153 (2024-10)
AgentSquare: Automatic LLM Agent Search in Modular Design Space
get_prompt_reasoning 1 other paper
1 dated earlier
2410.06153 (2024-10)
AgentSquare: Automatic LLM Agent Search in Modular Design Space
get_prompt_tooluse 1 other paper
1 dated earlier
2410.06153 (2024-10)
AgentSquare: Automatic LLM Agent Search in Modular Design Space
is_func 1 other paper
1 dated earlier
2502.12018 (2025-02)
Atom of Thoughts for Markov LLM Test-Time Scaling
merge_dict 1 other paper
1 dated earlier
2502.12018 (2025-02)
Atom of Thoughts for Markov LLM Test-Time Scaling
min_max_norm 1 other paper
1 later
2606.02359 (2026-06)
MOC: Multi-Order Communication in LLM-based Multi-Agent Systems

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

Compared with individual agents, large language model based multi-agent systems have shown great capabilities consistently across diverse tasks, including code generation, mathematical reasoning, and planning, etc. Despite their impressive performance, the effectiveness and robustness of these systems heavily rely on their communication topology, which is often fixed or generated in a single step. This restricts fine-grained structural exploration and flexible composition, resulting in excessive token utilization on simple tasks while limiting capability on complicated tasks. To mitigate this challenge, we introduce RADAR, a redundancy-aware and query-adaptive generative framework that actively reduce communication overhead. Motivated by recent progress in conditional discrete graph diffusion models, we formulate communication topology design as a step-bystep generation process, guided by the effective size of the graph. Comprehensive experiments on six benchmarks demonstrate that RADAR consistently outperforms recent baselines, achieving higher accuracy, lower token consumption, and greater robustness across diverse scenarios. Our code and data are available at https:// github.com/cszhangzhen/RADAR.

For agents

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

get_harvested_code_for_paper("2605.09907")
get_code_for_paper("2605.09907")
have("2605.09907")

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

curl https://syntology.ai/api/ran/2605.09907.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.09907.svg)](https://syntology.ai/paper/2605.09907)

Connect an agent — have() is free.