SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.09844 · 2024

Jack of All Trades, Master of Some, a Multi-Purpose Transformer Agent

arXiv · PDF · Open in the Atlas

Code that ran

6 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 failures of our instrument; 5 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/2402.09844.json · badge: /api/ran/2402.09844.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
huggingface/jat canonical (classifier label)
classifier basis: census
6 ran · 5 no recorded run
FunctionStatusWhere it lives
compute_ce_loss Ran
ran; our contract draft was wrong, not the code
huggingface/jat/jat/modeling_jat.py
code served (permissive licence) · get_code("ce965930fa44e9b4")
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
import torch.nn.functional as F

torch.manual_seed(0)

# Define ARGS dict with all required parameters
ARGS = {
    "logits": torch.randn(2, 3, 4),  # (batch_size=2, max_seq_len=3, num_classes=4)
    "labels": torch.randint(0, 4, (2, 3)),  # (batch_size=2, max_seq_len=3)
    "mask": torch.tensor([[True, True, False], [True, False, False]], dtype=torch.bool),  # (batch_size=2, max_seq_len=3)
    "weights": torch.ones(2, 3),  # (batch_size=2, max_seq_len=3)
}
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
compute_mse_loss Ran
ran; our contract draft was wrong, not the code
huggingface/jat/jat/modeling_jat.py
code served (permissive licence) · get_code("9889e7f5f482c648")
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
import torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors with distinct dimensions
batch_size = 2
max_seq_len = 3
feature_dim = 4

# predicted and true tensors: (batch_size, max_seq_len, feature_dim)
predicted = torch.randn(batch_size, max_seq_len, feature_dim)
true = torch.randn(batch_size, max_seq_len, feature_dim)

# mask tensor: (batch_size, max_seq_len) - boolean
mask = torch.tensor([
    [True, True, False],
    [True, False, False]
])

# weights tensor: (batch_size, max_seq_len) - float
weights = torch.tensor([
    [1.0, 0.5, 0.0],
    [1.0, 0.0, 0.0]
])

ARGS = {
    "predicted": predicted,
    "true": true,
    "mask": mask,
    "weights": weights
}
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
cyclic_expand_dim Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
huggingface/jat/jat/modeling_jat.py
code served (permissive licence) · get_code("50546bcb188a6472")
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 3, dtype family float, shape (2, 3, 7) 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 a small input tensor with shape (batch_size=2, seq_len=3, last_dim=4)
tensor = torch.tensor([
    [[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12]],
    [[13, 14, 15, 16], [17, 18, 19, 20], [21, 22, 23, 24]]
], dtype=torch.float32)

# The expanded_dim_size should be larger than the original last dimension (4)
expanded_dim_size = 7

ARGS = {
    "tensor": tensor,
    "expanded_dim_size": expanded_dim_size
}
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 3, dtype family float, shape (2, 3, 7) — shape (2, 3, 7) 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
pad Ran
ran; our contract draft was wrong, not the code
huggingface/jat/jat/processing_jat.py
code served (permissive licence) · get_code("a09e847bcb3917cc")
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

torch.manual_seed(0)

# Construct the encoding dictionary with sequences of lists
# The function expects encoding to be a dict where values are lists of sequences (lists)
encoding = {
    'feature1': [[1, 2], [3, 4, 5]],
    'feature2': [[10, 20, 30], [40]]
}

target_length = 4

ARGS = {
    'encoding': encoding,
    'target_length': target_length
}
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 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
to_tensor Ran
ran, honoured the contract we drafted
huggingface/jat/jat/processing_jat.py
code served (permissive licence) · get_code("cdde81732e084854")
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. 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
import numpy as np
import warnings

torch.manual_seed(0)

# Test case 1: Simple numpy array
x1 = np.array([1.0, 2.0, 3.0, 4.0])

# Test case 2: List of numpy arrays
x2 = [np.array([1.0, 2.0]), np.array([3.0, 4.0])]

# Test case 3: List of tensors
x3 = [torch.tensor([1.0, 2.0]), torch.tensor([3.0, 4.0])]

# Test case 4: Nested list structure
x4 = [[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]]

# Test case 5: Tuple of numpy arrays
x5 = (np.array([1.0, 2.0, 3.0]), np.array([4.0, 5.0, 6.0]))

# Test case 6: 2D numpy array
x6 = np.array([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])

# Use the simplest test case - a list of numpy arrays which tests the recursive path
ARGS = {
    'x': [np.array([1.0, 2.0, 3.0, 4.0]), np.array([5.0, 6.0, 7.0, 8.0])]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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
truncate Ran
ran; our contract draft was wrong, not the code
huggingface/jat/jat/processing_jat.py
code served (permissive licence) · get_code("92702d141fa25fd0")
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

torch.manual_seed(0)

# Construct encoding dictionary with multiple sequences of varying lengths
encoding = {
    'feature1': [[1, 2, 3, 4, 5], [6, 7, 8, 9, 10]],
    'feature2': [['a', 'b', 'c', 'd'], ['e', 'f', 'g']]
}

# Define ARGS dictionary with all parameters
ARGS = {
    'encoding': encoding,
    'max_length': 3,
    'truncation_side': 'right',
    'preserve': False
}
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 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
interleave_datasets Not yet run
unverified (harvested, no recorded run)
huggingface/jat/jat/utils_interleave_datasets.py
code served (permissive licence) · get_code("9c1882631bb2a9e9")
How it ran

no invocation recorded.

make_atari Not yet run
unverified (harvested, no recorded run)
huggingface/jat/jat/eval/rl/core.py
code served (permissive licence) · get_code("ff3946a18bfa4ec9")
How it ran

no invocation recorded.

make_babyai Not yet run
unverified (harvested, no recorded run)
huggingface/jat/jat/eval/rl/core.py
code served (permissive licence) · get_code("6166315cdf0375e4")
How it ran

no invocation recorded.

make_metaworld Not yet run
unverified (harvested, no recorded run)
huggingface/jat/jat/eval/rl/core.py
code served (permissive licence) · get_code("a5062b49b9e93ada")
How it ran

no invocation recorded.

normalize Not yet run
unverified (harvested, no recorded run)
huggingface/jat/jat/utils.py
code served (permissive licence) · get_code("4f68b1d9162757de")
How it ran

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 11 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 11 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

The search for a general model that can operate seamlessly across multiple domains remains a key goal in machine learning research. The prevailing methodology in Reinforcement Learning (RL) typically limits models to a single task within a unimodal framework, a limitation that contrasts with the broader vision of a versatile, multi-domain model. In this paper, we present Jack of All Trades (JAT), a transformer-based model with a unique design optimized for handling sequential decision-making tasks and multi-modal data types. The JAT model demonstrates its robust capabilities and versatility by achieving strong performance on very different RL benchmarks, along with promising results on Computer Vision (CV) and Natural Language Processing (NLP) tasks, all using a single set of weights. The JAT model marks a significant step towards more general, cross-domain AI model design, and notably, it is the first model of its kind to be fully open-sourced at https://huggingface.co/jat-project/jat, including a pioneering general-purpose dataset.

For agents

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

get_harvested_code_for_paper("2402.09844")
get_code_for_paper("2402.09844")
have("2402.09844")

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

curl https://syntology.ai/api/ran/2402.09844.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.09844.svg)](https://syntology.ai/paper/2402.09844)

Connect an agent — have() is free.