SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.21177 · 2026

ChunkFT: Byte-Streamed Optimization for Memory-Efficient Full Fine-Tuning

Qian Li, Shi Feng, Mengjie Zhao, Hinrich Schütze, Ercong Nie, Mingyang Wang, Daling Wang, Yongkang Liu, Feiliang Ren, Zijing Wang

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 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; 5 failures of our instrument; 2 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2605.21177.json · badge: /api/ran/2605.21177.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
misonsky/chunk canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
FunctionStatusWhere it lives
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
misonsky/chunk/models/modeling_llama.py
pointer only (licence: NONE) · get_code("3c76e52815c5401d")
How it ran

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

apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
misonsky/chunk/models/modeling_llama.py
pointer only (licence: NONE); the stored copy get_code serves, from jlamprou/Infini-Attention, is pointer only (licence: NONE) · get_code("0fe82a947dc39b42")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, misonsky/chunk 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_precision_recall Ran
ran, honoured the contract we drafted
misonsky/chunk/metrics/cuad/compute_score.py
pointer only (licence: NONE) · get_code("f69a366a76c79dcd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, misonsky/chunk 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_jaccard Ran
ran, honoured the contract we drafted
misonsky/chunk/metrics/cuad/compute_score.py
pointer only (licence: NONE) · get_code("c50aa95f98c19e6e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, misonsky/chunk 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_torch_cuda_memory_stats Ran
ran; our contract draft was wrong, not the code
misonsky/chunk/chnk/trainer.py
pointer only (licence: NONE) · get_code("8bd568d7e985846c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, misonsky/chunk 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_answer Ran
ran; our contract draft was wrong, not the code
misonsky/chunk/metrics/cuad/compute_score.py
pointer only (licence: NONE); the stored copy get_code serves, from borealisai/radar-decoding, is pointer only (licence: LGPL-3.0) · get_code("e7e75981cb464788")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, misonsky/chunk 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
misonsky/chunk/models/modeling_llama.py
pointer only (licence: NONE) · get_code("e03d53ba9d4f9ae5")
How it ran

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

postprocess_qa_predictions Not yet run
unverified (harvested, no recorded run)
misonsky/chunk/chnk/utils_qa.py
pointer only (licence: NONE) · get_code("ac56d3e7b79dbc18")
How it ran

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

postprocess_qa_predictions_with_beam_search Not yet run
unverified (harvested, no recorded run)
misonsky/chunk/chnk/utils_qa.py
pointer only (licence: NONE) · get_code("8c51f5aaf2f659e2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, misonsky/chunk 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 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 3 the same month, 4 later) and 3 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
340 dated earlier, 5 the same month, 12 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
repeat_kv 243 other papers
228 dated earlier, 4 the same month, 9 later, 2 undated · 22 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
2305.17888 (2023-05)
LLM-QAT: Data-Free Quantization Aware Training for Large Language Models
and 240 more not named here
normalize_answer 133 other papers
130 dated earlier, 2 later, 1 undated · 1 only as an assembly brick
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 130 more not named here
apply_rotary_pos_emb 75 other papers
73 dated earlier, 1 the same month, 1 later · 2 only as an assembly brick
2309.11499 (2023-09)
DreamLLM: Synergistic Multimodal Comprehension and Creation
2312.02051 (2023-12)
TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding
2312.03414 (2023-12)
Compressed Context Memory For Online Language Model Interaction
and 72 more not named here
postprocess_qa_predictions_with_beam_search 5 other papers
5 dated earlier
2106.09685 (2021-06)
LoRA: Low-Rank Adaptation of Large Language Models
2205.01068 (2022-05)
OPT: Open Pre-trained Transformer Language Models
2023.acl-short.31 (2023 (year only))
PLUE: Language Understanding Evaluation Benchmark for Privacy Policies in English
and 2 more not named here
postprocess_qa_predictions 4 other papers
4 dated earlier
2106.09685 (2021-06)
LoRA: Low-Rank Adaptation of Large Language Models
2023.acl-short.31 (2023 (year only))
PLUE: Language Understanding Evaluation Benchmark for Privacy Policies in English
2310.14709 (2023-10)
Once Upon a $\textit{Time}$ in $\textit{Graph}$: Relative-Time Pretraining for Complex Temporal Reasoning
and 1 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

This work presents CHUNKFT, a memory-efficient fine-tuning framework that reformulates full-parameter fine-tuning around a dynamically activated working set. CHUNKFT enables gradient computation for arbitrary sub-tensors without modifying the network architecture, providing an algorithmic foundation for optimizing arbitrary sub-networks while avoiding standard dense gradient computation. We provide a theoretical convergence analysis of CHUNKFT in the deterministic setting. Empirically, we apply CHUNKFT to fine-tune Llama 3-8B and Llama 3-70B using a single RTX 4090-24GB GPU and 2× H800-80GB GPUs, respectively. Full-parameter fine-tuning of a 7B model with a 1K input length requires only 13.72GB of GPU memory. The results demonstrate the effectiveness of CHUNKFT in memory usage, running time, and optimization quality. Moreover, downstream evaluations on language understanding, mathematical reasoning, and MT-Bench show that CHUNKFT consistently outperforms existing memory-efficient baselines. Notably, CHUNKFT achieves performance comparable to, and in some cases exceeding, full-parameter fine-tuning. Our repository is on https://github.com/misonsky/chunk.

For agents

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

get_harvested_code_for_paper("2605.21177")
get_code_for_paper("2605.21177")
have("2605.21177")

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

curl https://syntology.ai/api/ran/2605.21177.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.21177.svg)](https://syntology.ai/paper/2605.21177)

Connect an agent — have() is free.