SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.18005 · 2026

Flow-based Extremal Mathematical Structure Discovery

Gergely Bérczi, Baran Hashemi, Jonas Klüver

arXiv · PDF · Open in the Atlas

Code that ran

17 of 18 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; 5 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 11 failures of our instrument; 1 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/2601.18005.json · badge: /api/ran/2601.18005.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
berczig/FlowBoost canonical (classifier label)
classifier basis: census
17 ran · 1 no recorded run
FunctionStatusWhere it lives
all_triangle_areas Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/heilbronn_square/flow_matching_heilbronn.py
pointer only (licence: NONE) · get_code("391da1089de662c5")
How it ran

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

circle_wall_penalty Ran
ran on a synthesized input
berczig/FlowBoost/flow_boost/circles_in_square/flow_matching_sumradii.py
pointer only (licence: NONE) · get_code("2bff1ffa367143da")
How it ran

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

clamp_unit_square Ran
ran; our contract draft was wrong, not the code
berczig/FlowBoost/flow_boost/heilbronn_square/flow_matching_heilbronn.py
pointer only (licence: NONE) · get_code("8e2adb27913602ba")
How it ran

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

compute_metrics Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/heilbronn_square/plot_data_heilbronn.py
pointer only (licence: NONE) · get_code("2f705b01f997764b")
How it ran

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

compute_metrics Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/heilbronn_square/plot_data_points.py
pointer only (licence: NONE) · get_code("15fc8a32d3849637")
How it ran

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

compute_sum_radii Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/circles_in_square/plot_data_sumradii.py
pointer only (licence: NONE) · get_code("5bfedbd698572809")
How it ran

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

extract_bstart_bend Ran
ran, honoured the contract we drafted
berczig/FlowBoost/flow_boost/output/vmerge.py
pointer only (licence: NONE) · get_code("ca467f6c359703a2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, berczig/FlowBoost 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_min_area_and_triangles Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/heilbronn_square/sample_generation.py
pointer only (licence: NONE) · get_code("0a09310d7745fcc3")
How it ran

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

init_radii Ran
ran, honoured the contract we drafted
berczig/FlowBoost/flow_boost/circles_in_square/sample_generation.py
pointer only (licence: NONE) · get_code("5e7dba141ade472e")
How it ran

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

load_dataset Ran
ran; our contract draft was wrong, not the code
berczig/FlowBoost/flow_boost/circles_in_square/plot_data_sumradii.py
pointer only (licence: NONE) · get_code("96942d0b9f12bda2")
How it ran

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

load_dataset Ran
ran; our contract draft was wrong, not the code
berczig/FlowBoost/flow_boost/heilbronn_square/plot_data_heilbronn.py
pointer only (licence: NONE) · get_code("c0a6c1d41247a2b3")
How it ran

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

load_dataset Ran
ran; our contract draft was wrong, not the code
berczig/FlowBoost/flow_boost/heilbronn_square/plot_data_points.py
pointer only (licence: NONE) · get_code("cfbfce6523b26b63")
How it ran

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

sample_t Ran
ran, honoured the contract we drafted
berczig/FlowBoost/flow_boost/circles_in_square/flow_matching_sumradii.py
pointer only (licence: NONE) · get_code("0d73bc22ded8a72e")
How it ran

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

sample_uniform_centers Ran
ran, honoured the contract we drafted
berczig/FlowBoost/flow_boost/circles_in_square/sample_generation.py
pointer only (licence: NONE) · get_code("cafea8dbef362e75")
How it ran

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

sample_uniform_points Ran
ran, honoured the contract we drafted
berczig/FlowBoost/flow_boost/heilbronn_square/sample_generation.py
pointer only (licence: NONE) · get_code("91481f0658ae85d9")
How it ran

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

sample_uniform_x1_like Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/heilbronn_square/flow_matching_heilbronn.py
pointer only (licence: NONE) · get_code("903d7d4685ff1fda")
How it ran

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

validate_circle_samples Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
berczig/FlowBoost/flow_boost/circles_in_square/flow_matching_sumradii_centers_only.py
pointer only (licence: NONE) · get_code("bc905e1b5e297d92")
How it ran

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

circle_pair_penalty Not yet run
unverified (harvested, no recorded run)
berczig/FlowBoost/flow_boost/circles_in_square/flow_matching_sumradii.py
pointer only (licence: NONE) · get_code("58a710501d5d8ba3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, berczig/FlowBoost 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 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 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 18 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

The discovery of extremal structures in mathematics requires navigating vast and nonconvex landscapes where analytical methods offer little guidance and brute-force search becomes intractable. We introduce FlowBoost, a closed-loop generative framework that learns to discover rare and extremal combinatorial structures by combining three components: (i) a geometry-aware conditional flow-matching model that learns to sample high-quality configurations, (ii) reward-guided policy optimization with action exploration that directly optimizes the generation process toward the objective while maintaining diversity, and (iii) stochastic local search (Stochastic Relaxation with Perturbations) for both training-data generation and final refinement. Unlike prior open-loop approaches, such as PatternBoost [1] that either retrains on filtered discrete samples, or AlphaEvolve [2] where they rely on frozen Large Language Models (LLMs) as evolutionary mutation operators, FlowBoost enforces geometric feasibility during sampling, and propagates reward signal directly into the generative model, closing the optimization loop and requiring much smaller training sets and shorter training times, and reducing the required outer-loop iterations by orders of magnitude, while eliminating dependence on LLMs. We demonstrate the framework on four geometric optimization problems: sphere packing in hypercubes, circle packing maximizing sum of radii, the Heilbronn triangle problem, and star discrepancy minimization. In several cases, FlowBoost discovers configurations that match or exceed the best known results. For circle packings, we improve the best known lower bounds, surpassing the LLM-based system AlphaEvolve while using substantially fewer computational resources. For the Heilbronn problem, FlowBoost improves the minimum triangle area approaching the best known numerical values. For sphere packing in dimension 12, our method finds configurations denser than those produced by classical heuristics. To our knowledge, FlowBoost is the first systematic application of flow-based generative models with Reinforcement Learning to extremal structure discovery in pure mathematics. We term this paradigm de novo mathematical structure design. Our results demonstrate that lightweight, domain-specific generative models, when equipped with geometric inductive biases and reward-guided learning, can match or exceed the performance of LLM-based systems at a fraction of the computational cost. The code is available at https://github.com/berczig/FlowBoost.

For agents

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

get_harvested_code_for_paper("2601.18005")
get_code_for_paper("2601.18005")
have("2601.18005")

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

curl https://syntology.ai/api/ran/2601.18005.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.18005.svg)](https://syntology.ai/paper/2601.18005)

Connect an agent — have() is free.