SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2606.03465 · 2026

Rethinking the Role of Tensor Decompositions in Post-Training LLM Compression

Aleksandr Beznosikov, Vladimir Aletov, Gleb Molodtsov, Artur Zagitov, Alexander Miasnikov, Maxim Krutikov, Nail Bashirov, Artem Tsedenov

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 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; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 2 no recorded run), measured 2026-11-21; 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/2606.03465.json · badge: /api/ran/2606.03465.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
brain-lab-research/TT-LLM canonical (classifier label)
classifier basis: census
11 ran · 2 no recorded run
FunctionStatusWhere it lives
factor_int_balanced Ran
ran, honoured the contract we drafted
brain-lab-research/TT-LLM/src/tt_llm/tensor_utils.py
pointer only (licence: NONE) · get_code("245e53204100b72c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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_module_by_name Ran
ran; our contract draft was wrong, not the code
brain-lab-research/TT-LLM/src/tt_llm/llama_tt.py
pointer only (licence: NONE) · get_code("54053c7ed589cd79")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

infer_input_device Ran
ran; our contract draft was wrong, not the code
brain-lab-research/TT-LLM/src/tt_llm/superweights.py
pointer only (licence: NONE) · get_code("cd3d15ec465535aa")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

mixed_radix_unravel Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
brain-lab-research/TT-LLM/src/tt_llm/tt_superweights.py
pointer only (licence: NONE) · get_code("9b4e12658aeba2af")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

normalize_modes Ran
ran; our contract draft was wrong, not the code
brain-lab-research/TT-LLM/src/tt_llm/tensor_utils.py
pointer only (licence: NONE) · get_code("167ead9d0f467ae9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

prime_factors Ran
ran, honoured the contract we drafted
brain-lab-research/TT-LLM/src/tt_llm/tensor_utils.py
pointer only (licence: NONE) · get_code("dba3ae25db413cf8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

prune Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
brain-lab-research/TT-LLM/src/tucker_llm/laser/matrix_utils.py
pointer only (licence: NONE) · get_code("dfd3f0f25b0b9ad2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

rank Ran
ran, honoured the contract we drafted
brain-lab-research/TT-LLM/src/tucker_llm/laser/matrix_utils.py
pointer only (licence: NONE) · get_code("15557c4657639788")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

resolve_module Ran
ran; our contract draft was wrong, not the code
brain-lab-research/TT-LLM/src/tt_llm/superweights.py
pointer only (licence: NONE) · get_code("a09068db08c267a5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

resolve_transformer_layers Ran
ran; our contract draft was wrong, not the code
brain-lab-research/TT-LLM/src/tt_llm/superweights.py
pointer only (licence: NONE) · get_code("3968f618f7a63eb3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

sorted_mat Ran
ran, honoured the contract we drafted
brain-lab-research/TT-LLM/src/tucker_llm/laser/matrix_utils.py
pointer only (licence: NONE) · get_code("349808d7ef472b90")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

projection_mode_map Not yet run
unverified (harvested, no recorded run)
brain-lab-research/TT-LLM/src/tt_llm/llama_tt.py
pointer only (licence: NONE) · get_code("ff1ca01f4359bcdd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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.

replace_llama_ffn_with_tt Not yet run
unverified (harvested, no recorded run)
brain-lab-research/TT-LLM/src/tt_llm/llama_tt.py
pointer only (licence: NONE) · get_code("b0da99c1a090af3e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, brain-lab-research/TT-LLM 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 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 later) and 10 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
prune 1 other paper
1 dated earlier
2312.13558 (2023-12)
The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction
rank 1 other paper
1 dated earlier
2312.13558 (2023-12)
The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction
sorted_mat 1 other paper
1 dated earlier
2312.13558 (2023-12)
The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank Reduction

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Post-training compression is essential for deploying large language models (LLMs) under tight resource constraints. Tensor decompositions have emerged as a promising direction, offering compact parameterizations well suited to Transformer weight structures. However, existing studies evaluate these methods in narrow settings, leaving unclear whether tensorization is effective at large-scale deployment. We systematically evaluate tensor compression across dense and MoE architectures, establishing performance trade-offs grounded in both empirical analysis and theoretical analysis. We identify a fundamental mismatch between the shared subspaces assumed by tensor decompositions and the heterogeneous representations learned by modern LLMs, thereby delineating their practical limits and clarifying their viable role in large-scale deployment. The code is available at https://github.com/brain-lab-research/TT-LLM.

For agents

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

get_harvested_code_for_paper("2606.03465")
get_code_for_paper("2606.03465")
have("2606.03465")

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

curl https://syntology.ai/api/ran/2606.03465.json

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

[![Syntology run record](https://syntology.ai/api/ran/2606.03465.svg)](https://syntology.ai/paper/2606.03465)

Connect an agent — have() is free.