SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.22681 · 2026

Accelerating LLM Pre-Training through Flat-Direction Dynamics Enhancement

Kun Yuan, Xue Wang, Mingze Wang, Zaiwen Wen, Rizhen Hu, Mou Sun, Shuchen Zhu

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 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/2602.22681.json · badge: /api/ran/2602.22681.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
SHUCHENZHU/LITE canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SHUCHENZHU/LITE/peft_pretraining/modeling_llama.py
pointer only (licence: NONE) · get_code("f725bc2d76076485")
How it ran

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

beta_scheduler Ran
ran, honoured the contract we drafted
SHUCHENZHU/LITE/muonlite.py
pointer only (licence: NONE) · get_code("25f37cd0df62256a")
How it ran

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

cosine_lr_schedule Ran
ran, honoured the contract we drafted
SHUCHENZHU/LITE/muonlite.py
pointer only (licence: NONE) · get_code("f28cdb4832ce7cf3")
How it ran

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

get_cyclical_cosine_schedule_with_min_lr Ran
ran; our contract draft was wrong, not the code
SHUCHENZHU/LITE/peft_pretraining/training_utils.py
pointer only (licence: NONE) · get_code("0311c26d2bf832a0")
How it ran

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

get_wsd_schedule_with_min_lr Ran
ran; our contract draft was wrong, not the code
SHUCHENZHU/LITE/peft_pretraining/training_utils.py
pointer only (licence: NONE) · get_code("d277e7e92ab2f75b")
How it ran

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

rotate_half Ran
ran; our contract draft was wrong, not the code
SHUCHENZHU/LITE/peft_pretraining/modeling_llama.py
pointer only (licence: NONE) · get_code("b99eea6376d1e212")
How it ran

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

get_scheculer Not yet run
unverified (harvested, no recorded run)
SHUCHENZHU/LITE/peft_pretraining/training_utils.py
pointer only (licence: NONE) · get_code("59f17ef66274a3ca")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, SHUCHENZHU/LITE 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 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 3 later) and 4 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
rotate_half 360 other papers
324 dated earlier, 6 the same month, 27 later, 3 undated · 30 only as an assembly brick
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
2203.08913 (2022-03)
Memorizing Transformers
2302.13971 (2023-02)
LLaMA: Open and Efficient Foundation Language Models
and 357 more not named here
apply_rotary_pos_emb 83 other papers
79 dated earlier, 1 the same month, 3 later · 6 only as an assembly brick
2306.15595 (2023-06)
Extending Context Window of Large Language Models via Positional Interpolation
2307.05695 (2023-07)
ReLoRA: High-Rank Training Through Low-Rank Updates
2307.06945 (2023-07)
In-context Autoencoder for Context Compression in a Large Language Model
and 80 more not named here
get_cyclical_cosine_schedule_with_min_lr 7 other papers
5 dated earlier, 2 later
2307.05695 (2023-07)
ReLoRA: High-Rank Training Through Low-Rank Updates
2406.02214 (2024-06)
SLTrain: a sparse plus low-rank approach for parameter and memory efficient pretraining
2407.08296 (2024-07)
Q-GaLore: Quantized GaLore with INT4 Projection and Layer-Adaptive Low-Rank Gradients
and 4 more not named here

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

Pre-training Large Language Models requires immense computational resources, making optimizer efficiency essential. The optimization landscape is highly anisotropic, with loss reduction driven predominantly by progress along flat directions. While matrix-based optimizers such as Muon and SOAP leverage fine-grained curvature information to outperform AdamW, their updates tend toward isotropy-relatively conservative along flat directions yet potentially aggressive along sharp ones. To address this limitation, we first establish a unified Riemannian Ordinary Differential Equation (ODE) framework that elucidates how common adaptive algorithms operate synergistically: the preconditioner induces a Riemannian geometry that mitigates illconditioning, while momentum serves as a Riemannian damping term that promotes convergence. Guided by these insights, we propose LITE, a generalized acceleration strategy that enhances training dynamics by applying larger Hessian damping coefficients and learning rates along flat trajectories. Extensive experiments demonstrate that LITE significantly accelerates both Muon and SOAP across diverse architectures (Dense, MoE), parameter scales (130M-1.3B), datasets (C4, Pile), and learning-rate schedules (cosine, warmup-stable-decay). Theoretical analysis confirms that LITE facilitates faster convergence along flat directions in anisotropic landscapes, providing a principled approach to efficient LLM pre-training. The code is available at https://github.com/SHUCHENZHU/LITE.

For agents

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

get_harvested_code_for_paper("2602.22681")
get_code_for_paper("2602.22681")
have("2602.22681")

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

curl https://syntology.ai/api/ran/2602.22681.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.22681.svg)](https://syntology.ai/paper/2602.22681)

Connect an agent — have() is free.