SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.25976 · 2026

Second-Order, First-Class: A Composable Stack for Curvature-Aware Training

Mikalai Korbit, Mario Zanon

arXiv · PDF · Open in the Atlas

Code that ran

1 of 6 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure 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/2603.25976.json · badge: /api/ran/2603.25976.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
cor3bit/somax canonical (classifier label)
classifier basis: census
1 ran · 5 no recorded run
FunctionStatusWhere it lives
register Ran
ran; our contract draft was wrong, not the code
cor3bit/somax/src/somax/presets.py
code served (permissive licence) · get_code("bf106c4e3cfa927f")
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 function. The image it ran in was not recorded. If your copy returns the same type 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
import torch.nn as nn
from typing import Any, Callable

# Initialize the global registry that the decorator uses
_REG = {}
_DESC = {}

# Define arguments for the register function
ARGS = {
    'name': 'test_preset',
    'desc': 'A test preset factory function'
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
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_curvature Not yet run
unverified (harvested, no recorded run)
cor3bit/somax/src/somax/specs.py
code served (permissive licence) · get_code("fe6c04bf91489f8c")
How it ran

no invocation recorded.

build_damping Not yet run
unverified (harvested, no recorded run)
cor3bit/somax/src/somax/specs.py
code served (permissive licence) · get_code("ad9ad261d72a6a69")
How it ran

no invocation recorded.

default Not yet run
unverified (harvested, no recorded run)
cor3bit/somax/src/somax/metrics.py
code served (permissive licence) · get_code("bee5b61d1f422153")
How it ran

no invocation recorded.

describe Not yet run
unverified (harvested, no recorded run)
cor3bit/somax/src/somax/presets.py
code served (permissive licence) · get_code("fe60c4cd5931c27e")
How it ran

no invocation recorded.

make Not yet run
unverified (harvested, no recorded run)
cor3bit/somax/src/somax/presets.py
code served (permissive licence) · get_code("d4d746521afc0313")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 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 6 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

Second-order methods promise improved stability and faster convergence, yet they remain underused due to implementation overhead, tuning brittleness, and the lack of composable APIs. We introduce Somax, a composable Optax-native stack that treats curvature-aware training as a single JIT-compiled step governed by a static plan. Somax exposes first-class modules -curvature operators, estimators, linear solvers, preconditioners, and damping policies -behind a single step interface and composes with Optax by applying standard gradient transformations (e.g., momentum, weight decay, schedules) to the computed direction. This design makes typically hidden choices explicit and swappable. Somax separates planning from execution: it derives a static plan (including cadences) from module requirements, then runs the step through a specialized execution path that reuses intermediate results across modules. We report systemoriented ablations showing that (i) composition choices materially affect scaling behavior and time-to-accuracy, and (ii) planning reduces per-step overhead relative to unplanned composition with redundant recomputation.

For agents

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

get_harvested_code_for_paper("2603.25976")
get_code_for_paper("2603.25976")
have("2603.25976")

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

curl https://syntology.ai/api/ran/2603.25976.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.25976.svg)](https://syntology.ai/paper/2603.25976)

Connect an agent — have() is free.