SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.18865 · 2024

Analyzing and Reducing Catastrophic Forgetting in Parameter Efficient Tuning

arXiv · PDF · Open in the Atlas

Code that ran

9 of 11 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; 5 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2402.18865.json · badge: /api/ran/2402.18865.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
which47/llmcl canonical (classifier label)
classifier basis: census
9 ran · 2 no recorded run
FunctionStatusWhere it lives
concat_inputs Ran
ran; our contract draft was wrong, not the code
which47/llmcl/method/ER.py
pointer only (licence: NONE) · get_code("18b8186589457eb4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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_joint_datasets Ran
ran; our contract draft was wrong, not the code
which47/llmcl/get_dataset.py
pointer only (licence: NONE) · get_code("dbea6b5c8c82bfb1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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_res Ran
ran, honoured the contract we drafted
which47/llmcl/evaluation.py
pointer only (licence: NONE) · get_code("f6fd640bd5fb5f16")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

l2_normalize Ran
ran; our contract draft was wrong, not the code
which47/llmcl/method/L2P.py
pointer only (licence: NONE) · get_code("d16edbaed144f08d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

list_strings Ran
ran; our contract draft was wrong, not the code
which47/llmcl/utils/arg_configs.py
pointer only (licence: NONE) · get_code("9a37fcbe00419582")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

reservoir Ran
ran, honoured the contract we drafted
which47/llmcl/method/ILORA.py
pointer only (licence: NONE) · get_code("7e522e4817696b98")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

reservoir Ran
ran, honoured the contract we drafted
which47/llmcl/method/ER.py
pointer only (licence: NONE) · get_code("04102179b159097d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

ring Ran
ran, honoured the contract we drafted
which47/llmcl/method/ILORA.py
pointer only (licence: NONE); the stored copy get_code serves, from Kaile-Du/CSC, is pointer only (licence: NONE) · get_code("3ea09d9b15eac6ea")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

run_generation Ran
ran; our contract draft was wrong, not the code
which47/llmcl/inference.py
pointer only (licence: NONE) · get_code("68c73a68bb1fa2f3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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_datasets Not yet run
unverified (harvested, no recorded run)
which47/llmcl/get_dataset.py
pointer only (licence: NONE) · get_code("d90e6d4a6e725608")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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.

prepare_model_for_inference Not yet run
unverified (harvested, no recorded run)
which47/llmcl/inference.py
pointer only (licence: NONE) · get_code("4277a360496e0802")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, which47/llmcl 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 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 later) and 8 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
reservoir 13 other papers
9 dated earlier, 4 later · 7 only as an assembly brick
2010.05595 (2020-10)
Rethinking Experience Replay: a Bag of Tricks for Continual Learning
2110.06976 (2021-10)
Representational Continuity for Unsupervised Continual Learning
2201.12604 (2022-01)
Learning Fast, Learning Slow: A General Continual Learning Method based on Complementary Learning System
and 10 more not named here
ring 1 other paper
1 later
2403.12559 (2024-03)
Confidence Self-Calibration for Multi-Label Class-Incremental Learning

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

Existing research has shown that large language models (LLMs) exhibit remarkable performance in language understanding and generation. However, when LLMs are continuously fine-tuned on complex and diverse domain-specific downstream tasks, the inference performance on historical tasks decreases dramatically, which is known as a catastrophic forgetting problem. A trade-off needs to be kept between learning plasticity and memory stability. Plenty of existing works have explored strategies like memory replay, regularization and parameter isolation, but little is known about the geometric connection of various adjacent minima in the continual LLMs fine-tuning scenarios. In this work, we investigate the geometric connections of different minima through the lens of mode connectivity, which means different minima can be connected by a low-loss valley. Through extensive experiments, we uncover the mode connectivity phenomenon in the LLMs continual learning scenario and find that it can strike a balance between plasticity and stability. Building upon these findings, we propose a simple yet effective method called Interpolation-based LoRA (I-LoRA), which constructs a dual-memory experience replay framework based on LoRA parameter interpolations. Extensive experiments and analysis on eight domain-specific CL benchmarks demonstrate that I-LoRA consistently show significant improvement over the previous state-of-the-art approaches with up to $11\%$ performance gains, providing a strong baseline and insights for future research on the large language model continual learning problem. Our code is available at \url{https://github.com/which47/LLMCL}.

For agents

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

get_harvested_code_for_paper("2402.18865")
get_code_for_paper("2402.18865")
have("2402.18865")

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

curl https://syntology.ai/api/ran/2402.18865.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.18865.svg)](https://syntology.ai/paper/2402.18865)

Connect an agent — have() is free.