SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2311.05348 · 2023

u-LLaVA: Unifying Multi-Modal Tasks via Large Language Model

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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 1 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/2311.05348.json · badge: /api/ran/2311.05348.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
OPPOMKLab/u-LLaVA canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
bbox_iou Ran
ran; our contract draft was wrong, not the code
OPPOMKLab/u-LLaVA/evaluation/tools.py
code served (permissive licence) · get_code("8b0261c8ffb51491")
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 dict. 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
from torchvision.ops import box_iou

torch.manual_seed(0)

# Create small bounding box tensors
# Format: [x1, y1, x2, y2] for each box
# pred_boxes: 2 predicted boxes
pred_boxes = torch.tensor([
    [0.1, 0.1, 0.5, 0.5],
    [0.2, 0.2, 0.6, 0.6]
], dtype=torch.float32)

# target_boxes: 2 target boxes
target_boxes = torch.tensor([
    [0.1, 0.1, 0.5, 0.5],
    [0.3, 0.3, 0.7, 0.7]
], dtype=torch.float32)

ARGS = {
    'pred_boxes': pred_boxes,
    'target_boxes': target_boxes
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
box_iou Ran
ran, honoured the contract we drafted
OPPOMKLab/u-LLaVA/models/loss.py
code served (permissive licence) · get_code("ce749f837424dd8c")
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, ndim 3, 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
from torchvision.ops.boxes import box_area

torch.manual_seed(0)

# Create two sets of bounding boxes in format [x1, y1, x2, y2]
# boxes1: 2 boxes
# boxes2: 3 boxes
boxes1 = torch.tensor([
    [0.0, 0.0, 2.0, 2.0],
    [1.0, 1.0, 3.0, 3.0]
], dtype=torch.float32)

boxes2 = torch.tensor([
    [0.0, 0.0, 1.0, 1.0],
    [1.5, 1.5, 2.5, 2.5],
    [2.0, 2.0, 4.0, 4.0]
], dtype=torch.float32)

ARGS = {
    "boxes1": boxes1,
    "boxes2": boxes2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, dtype family float — a tuple 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
dice_loss Ran
ran; our contract draft was wrong, not the code
OPPOMKLab/u-LLaVA/models/loss.py
code served (permissive licence) · get_code("faf023207d110281")
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 Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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)

# Create small tensors for inputs and targets
# inputs: arbitrary shape float tensor with predictions
# targets: same shape as inputs, binary labels (0 or 1)
batch_size = 2
height = 4
width = 3

inputs = torch.randn(batch_size, height, width, dtype=torch.float32)
targets = torch.randint(0, 2, (batch_size, height, width), dtype=torch.float32)

# num_masks: number of masks, used for normalization
num_masks = 2.0

# scale: scaling factor for normalization
scale = 1000

# eps: small epsilon value for numerical stability
eps = 1e-06

ARGS = {
    "inputs": inputs,
    "targets": targets,
    "num_masks": num_masks,
    "scale": scale,
    "eps": eps,
}
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 Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
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
find_linear_layers Ran
ran; our contract draft was wrong, not the code
OPPOMKLab/u-LLaVA/train_ullava.py
code served (permissive licence) · get_code("3335a08070d5bb43")
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)

# Build a minimal model with Linear layers and various module names
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        # These should be included (match lora_target_modules)
        self.attention_layer = nn.Linear(4, 3)
        self.mlp_dense = nn.Linear(3, 2)
        self.query_proj = nn.Linear(2, 4)
        
        # These should be excluded (contain vision/det/seg patterns)
        self.vision_encoder = nn.Linear(4, 3)
        self.vision_projector = nn.Linear(3, 2)
        self.seg_projector = nn.Linear(2, 4)
        self.visual_model = nn.Linear(4, 3)
        self.det_projector = nn.Linear(3, 2)
        self.det_decoder = nn.Linear(2, 4)
        
        # Nested structure to test named_modules
        self.block = nn.Sequential(
            nn.Linear(4, 3),
            nn.ReLU()
        )
        self.block_attention = nn.Linear(3, 2)
        
    def forward(self, x):
        return x

model = SimpleModel()

# Define lora_target_modules - these patterns should match layer names to include them
lora_target_modules = ['attention', 'mlp', 'query', 'block']

ARGS = {
    'model': model,
    'lora_target_modules': lora_target_modules
}
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
generalized_box_iou Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
OPPOMKLab/u-LLaVA/models/loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from roymiles/vkd, is pointer only (licence: NONE) · get_code("dc8fb9b6fb48ac37")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from roymiles/vkd, and GitHub detected no licence in that repository 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.

to_cuda Ran
ran; our contract draft was wrong, not the code
OPPOMKLab/u-LLaVA/evaluation/tools.py
code served (permissive licence) · get_code("e7bb24cbb35466a5")
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 Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, ndim and shape 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)

# Create a small CPU tensor
tensor = torch.randn(2, 4)

# Define ARGS dict with all parameters
ARGS = {
    "tensor": tensor,
    "torch_type": 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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
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
intersectionAndUnionGPU Not yet run
unverified (harvested, no recorded run)
OPPOMKLab/u-LLaVA/evaluation/tools.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Cavendish518/SFPNet, is pointer only (licence: NOASSERTION) · get_code("136834d66336e6b9")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from Cavendish518/SFPNet, and GitHub could not classify the licence of that repository (NOASSERTION). 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, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 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
box_iou 24 other papers
13 dated earlier, 10 later, 1 undated · 5 only as an assembly brick
2103.12115 (2021-03)
End-to-End Trainable Multi-Instance Pose Estimation with Transformers
2104.00969 (2021-04)
TubeR: Tubelet Transformer for Video Action Detection
2112.05375 (2021-12)
Rethinking the Two-Stage Framework for Grounded Situation Recognition
and 21 more not named here
dice_loss 16 other papers
16 later · 1 only as an assembly brick
2312.02228 (2023-12)
PixelLM: Pixel Reasoning with Large Multimodal Model
2312.10103 (2023-12)
GSVA: Generalized Segmentation via Multimodal Large Language Models
2312.15715 (2023-12)
Segment Every Reference Object in Spatial and Temporal Spaces
and 13 more not named here
intersectionAndUnionGPU 9 other papers
5 dated earlier, 3 later, 1 same year, order not determinable
2108.05293 (2021-08)
Few-Shot Segmentation with Global and Local Contrastive Learning
2203.14508 (2022-03)
Stratified Transformer for 3D Point Cloud Segmentation
2210.06780 (2022-10)
Intermediate Prototype Mining Transformer for Few-Shot Semantic Segmentation
and 6 more not named here
generalized_box_iou 6 other papers
2 dated earlier, 4 later · 3 only as an assembly brick
2303.13076 (2023-03)
CORA: Adapting CLIP for Open-Vocabulary Detection with Region Prompting and Anchor Pre-Matching
2304.04742 (2023-04)
Detection Transformer with Stable Matching
2403.06213 (2024-03)
V k D : Improving Knowledge Distillation using Orthogonal Projections
and 3 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

Recent advancements in multi-modal large language models (MLLMs) have led to substantial improvements in visual understanding, primarily driven by sophisticated modality alignment strategies. However, predominant approaches prioritize global or regional comprehension, with less focus on fine-grained, pixel-level tasks. To address this gap, we introduce u-LLaVA, an innovative unifying multi-task framework that integrates pixel, regional, and global features to refine the perceptual faculties of MLLMs. We commence by leveraging an efficient modality alignment approach, harnessing both image and video datasets to bolster the model's foundational understanding across diverse visual contexts. Subsequently, a joint instruction tuning method with task-specific projectors and decoders for end-to-end downstream training is presented. Furthermore, this work contributes a novel mask-based multi-task dataset comprising 277K samples, crafted to challenge and assess the fine-grained perception capabilities of MLLMs. The overall framework is simple, effective, and achieves state-of-the-art performance across multiple benchmarks. We also make our model, data, and code publicly accessible at https://github.com/OPPOMKLab/u-LLaVA.

For agents

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

get_harvested_code_for_paper("2311.05348")
get_code_for_paper("2311.05348")
have("2311.05348")

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

curl https://syntology.ai/api/ran/2311.05348.json

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

[![Syntology run record](https://syntology.ai/api/ran/2311.05348.svg)](https://syntology.ai/paper/2311.05348)

Connect an agent — have() is free.