SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.05231 · ECCV · 2024

Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance

Yaowei Wang, Yong Xu, Zhipeng Zhang, Haibin Ling, Heng Fan, Liting Lin

arXiv · PDF · Open in the Atlas

Code that ran

3 of 5 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; 3 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/2403.05231.json · badge: /api/ran/2403.05231.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
LitingLin/LoRAT canonical (classifier label)
classifier basis: census
3 ran · 2 no recorded run
FunctionStatusWhere it lives
find_all_frozen_nn_linear_names Ran
ran; our contract draft was wrong, not the code
LitingLin/LoRAT/trackit/models/methods/LoRAT/funcs/vit_lora_utils.py
code served (permissive licence) · get_code("5a695ef0811cab18")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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

torch.manual_seed(0)

# Create a model with some frozen and some unfrozen Linear layers
class TestModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer1 = nn.Linear(4, 3)
        self.layer2 = nn.Linear(3, 2)
        self.layer3 = nn.Linear(2, 5)
        
        # Freeze layer1 and layer3
        self.layer1.weight.requires_grad = False
        self.layer1.bias.requires_grad = False
        self.layer3.weight.requires_grad = False
        self.layer3.bias.requires_grad = False
        
        # Keep layer2 unfrozen
        self.layer2.weight.requires_grad = True
        self.layer2.bias.requires_grad = True

model = TestModel()

ARGS = {
    "model": model
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
freeze_vit_backbone_ Ran
ran; our contract draft was wrong, not the code
LitingLin/LoRAT/trackit/models/methods/LoRAT/funcs/vit_backbone_freeze.py
code served (permissive licence) · get_code("4b916f2f46e88d85")
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 MinimalViT. 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 itertools import chain

torch.manual_seed(0)

# Create a minimal ViT-style model with the required attributes
class PatchEmbedding(nn.Module):
    def __init__(self):
        super().__init__()
        self.proj = nn.Linear(3, 4)
    
    def forward(self, x):
        return self.proj(x)

class TransformerBlock(nn.Module):
    def __init__(self):
        super().__init__()
        self.attn = nn.Linear(4, 4)
        self.mlp = nn.Linear(4, 4)
    
    def forward(self, x):
        return self.mlp(self.attn(x))

class MinimalViT(nn.Module):
    def __init__(self):
        super().__init__()
        self.patch_embed = PatchEmbedding()
        self.blocks = nn.Sequential(
            TransformerBlock(),
            TransformerBlock()
        )
        self.pos_embed = nn.Parameter(torch.randn(1, 8, 4))
        self.norm = nn.LayerNorm(4)
    
    def forward(self, x):
        x = self.patch_embed(x)
        x = self.blocks(x)
        x = self.norm(x)
        return x

model = MinimalViT()

ARGS = {
    "model": model,
    "freeze_out_norm": True,
}
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 MinimalViT — a MinimalViT output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MinimalViT 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
generate_LoRAT_sample_data Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LitingLin/LoRAT/trackit/models/methods/LoRAT/funcs/sample_data.py
code served (permissive licence) · get_code("c311b3c84e280fa0")
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 tuple. 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

torch.manual_seed(0)

ARGS = {
    "z_feat_size": (4, 3),
    "x_feat_size": (5, 6),
    "patch_size": (2, 2),
    "batch_size": 2,
    "device": torch.device("cpu"),
    "dtype": torch.float32,
}
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 tuple — a tuple output has no array shape of its own; none is 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
enable_lora_ Not yet run
unverified (harvested, no recorded run)
LitingLin/LoRAT/trackit/models/methods/LoRAT/funcs/vit_lora_utils.py
code served (permissive licence) · get_code("7dcc0654aaf67867")
How it ran

no invocation recorded.

state_dict_with_lora_meta_attributes Not yet run
unverified (harvested, no recorded run)
LitingLin/LoRAT/trackit/models/methods/LoRAT/funcs/vit_lora_utils.py
code served (permissive licence) · get_code("f6f3949d8af0fe7f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 5 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

Motivated by the Parameter-Efficient Fine-Tuning (PEFT) in large language models, we propose LoRAT, a method that unveils the power of larger Vision Transformers (ViT) for tracking within laboratorylevel resources. The essence of our work lies in adapting LoRA, a technique that fine-tunes a small subset of model parameters without adding inference latency, to the domain of visual tracking. However, unique challenges and potential domain gaps make this transfer not as easy as the first intuition. Firstly, a transformer-based tracker constructs unshared position embedding for template and search image. This poses a challenge for the transfer of LoRA, usually requiring consistency in the design when applied to the pre-trained backbone, to downstream tasks. Secondly, the inductive bias inherent in convolutional heads diminishes the effectiveness of parameter-efficient fine-tuning in tracking models. To overcome these limitations, we first decouple the position embeddings in transformer-based trackers into shared spatial ones and independent type ones. The shared embeddings, which describe the absolute coordinates of multi-resolution images (namely, the template and search images), are inherited from the pre-trained backbones. In contrast, the independent embeddings indicate the sources of each token and are learned from scratch. Furthermore, we design an anchor-free head solely based on a multilayer perceptron (MLP) to adapt PETR, enabling better performance with less computational overhead. With our design, 1) it becomes practical to train trackers with the ViT-g backbone on GPUs with only memory of 25.8GB (batch size of 16); 2) we reduce the training time of the L-224 variant from 35.0 to 10.8 GPU hours; 3) we improve the LaSOT SUC score from 0.703 to 0.742 with the L-224 variant; 4) we fast the inference speed of the L-224 variant from 52 to 119 FPS. Code and models are available at https://github.com/LitingLin/LoRAT.

For agents

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

get_harvested_code_for_paper("2403.05231")
get_code_for_paper("2403.05231")
have("2403.05231")

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

curl https://syntology.ai/api/ran/2403.05231.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.05231.svg)](https://syntology.ai/paper/2403.05231)

Connect an agent — have() is free.