SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1903.01602 · 2019

The Regretful Agent: Heuristic-Aided Navigation through Progress Estimation

arXiv · PDF · Open in the Atlas

Code that ran

10 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 7 have no contract verdict; 2 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/1903.01602.json · badge: /api/ran/1903.01602.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
chihyaoma/regretful-agent canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
CAVED123/regretful-agent community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
cacosandon/are-you-looking community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
FunctionStatusWhere it lives
build_mlp Ran
ran on a synthesized input
CAVED123/regretful-agent/tasks/R2R-pano/models/modules.py
code served (permissive licence) · get_code("91a412404a2ae592")
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 Sequential. 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

torch.manual_seed(0)

ARGS = {
    "input_dim": 4,
    "hidden_dims": [8, 16],
    "output_dim": 2,
    "use_batchnorm": True,
    "dropout": 0.5,
    "fc_bias": True,
    "relu": 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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
build_vocab Ran
ran; our contract draft was wrong, not the code
CAVED123/regretful-agent/tasks/R2R-pano/utils.py
code served (permissive licence) · get_code("19449a71349e2621")
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 json
import tempfile
import os

# Create temporary directory and mock data files
tmpdir = tempfile.mkdtemp()

# Mock R2R dataset files
train_data = [
    {
        "instructions": [
            "Walk forward and turn left at the corner",
            "Go straight ahead then turn right"
        ]
    },
    {
        "instructions": [
            "Walk forward slowly",
            "Turn around and go back"
        ]
    }
]

val_seen_data = [
    {
        "instructions": [
            "Move forward to the door"
        ]
    }
]

# Write mock data files
os.makedirs(os.path.join(tmpdir, 'tasks/R2R-pano/data'), exist_ok=True)

with open(os.path.join(tmpdir, 'tasks/R2R-pano/data/R2R_train.json'), 'w') as f:
    json.dump(train_data, f)

with open(os.path.join(tmpdir, 'tasks/R2R-pano/data/R2R_val_seen.json'), 'w') as f:
    json.dump(val_seen_data, f)

with open(os.path.join(tmpdir, 'tasks/R2R-pano/data/R2R_val_unseen.json'), 'w') as f:
    json.dump([], f)

with open(os.path.join(tmpdir, 'tasks/R2R-pano/data/R2R_test.json'), 'w') as f:
    json.dump([], f)

with open(os.path.join(tmpdir, 'tasks/R2R-pano/data/R2R_train_val_seen.json'), 'w') as f:
    json.dump([], f)

with open(os.path.join(tmpdir, 'tasks/R2R-pano/data/R2R_literal_speaker_data_augmentation_paths.json'), 'w') as f:
    json.dump([], f)

# Change to temp directory for file loading
os.chdir(tmpdir)

ARGS = {
    'splits': ['train', 'val_seen'],
    'min_count': 1,
    'start_vocab': ['<PAD>', '<START>', '<EOS>', '<UNK>']
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
build_vocab Ran
ran on a synthesized input
cacosandon/are-you-looking/tasks/R2R-pano/utils.py
code served (permissive licence) · get_code("beb5b1f67bf44b76")
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 tempfile
import json
import os

base_vocab = ['<PAD>', '<START>', '<EOS>', '<UNK>']

class MockOpts:
    def __init__(self, instructions_path, train_instructions_name, data_augmentation_name):
        self.instructions = instructions_path
        self.train_instructions = train_instructions_name
        self.path_data_augmentation = data_augmentation_name

tmpdir = tempfile.mkdtemp()

train_data = [
    {
        "instructions": [
            "go straight ahead and turn left",
            "walk forward then right",
        ]
    },
    {
        "instructions": [
            "move ahead and turn around",
            "go left at the corner",
        ]
    }
]

val_seen_data = [
    {
        "instructions": [
            "walk forward",
            "turn right",
        ]
    }
]

with open(os.path.join(tmpdir, "R2R_train.json"), "w") as f:
    json.dump(train_data, f)

with open(os.path.join(tmpdir, "R2R_val_seen.json"), "w") as f:
    json.dump(val_seen_data, f)

with open(os.path.join(tmpdir, "synthetic.json"), "w") as f:
    json.dump(train_data, f)

opts = MockOpts(
    instructions_path=tmpdir,
    train_instructions_name="train",
    data_augmentation_name="synthetic.json"
)

ARGS = {
    'splits': ['train'],
    'min_count': 1,
    'start_vocab': base_vocab,
    'opts': opts
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
count_rollback_success Ran
ran, honoured the contract we drafted
chihyaoma/regretful-agent/tasks/R2R-pano/trainer.py
code served (permissive licence) · get_code("b8a9ddd7f989b2b6")
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) and returned tuple, ndim 1, dtype family int; 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

torch.manual_seed(0)

# Construct valid arguments for count_rollback_success
# The function expects:
# - success_count: int
# - rollback_success_count: int
# - rollback_count: int
# - oscillating_success_count: int
# - oscillating_count: int
# - traj: list of trajectory dicts
# - error_margin: float (optional, default 3.0)

# Each trajectory dict needs:
# - 'path': list of tuples (viewpoint, _, _)
# - 'distance': list of floats

# Create sample trajectories
traj = [
    {
        'path': [
            ('viewpoint_0', 0, 0),
            ('viewpoint_1', 1, 1),
            ('viewpoint_0', 2, 2),
            ('viewpoint_2', 3, 3),
        ],
        'distance': [10.0, 8.0, 6.0, 2.0]
    },
    {
        'path': [
            ('viewpoint_a', 0, 0),
            ('viewpoint_b', 1, 1),
            ('viewpoint_c', 2, 2),
            ('viewpoint_b', 3, 3),
        ],
        'distance': [15.0, 12.0, 5.0, 1.5]
    },
    {
        'path': [
            ('viewpoint_x', 0, 0),
            ('viewpoint_y', 1, 1),
            ('viewpoint_z', 2, 2),
        ],
        'distance': [20.0, 18.0, 25.0]
    }
]

ARGS = {
    'success_count': 0,
    'rollback_success_count': 0,
    'rollback_count': 0,
    'oscillating_success_count': 0,
    'oscillating_count': 0,
    'traj': traj,
    'error_margin': 3.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family int — 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
create_mask Ran
ran on a synthesized input
CAVED123/regretful-agent/tasks/R2R-pano/models/modules.py
code served (permissive licence) · get_code("f3bcfc9ebf343e45")
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)

batchsize = 2
max_length = 4
length = torch.tensor([2, 3])

ARGS = {
    "batchsize": batchsize,
    "max_length": max_length,
    "length": 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 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
load_dataset Ran
ran on a synthesized input
cacosandon/are-you-looking/modify-subjects/utils.py
code served (permissive licence) · get_code("81d6dca31149202f")
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 json
import tempfile
import os

# Create temporary directory and mock data files
temp_dir = tempfile.mkdtemp()
tasks_dir = os.path.join(temp_dir, 'tasks', 'R2R-pano', 'data')
os.makedirs(tasks_dir, exist_ok=True)

# Create mock R2R data files
mock_data = {
    "instruction_id": "R2R_train_00000",
    "path_id": 0,
    "instructions": ["Go straight ahead"],
    "trajectory": [[0, 0, 0]],
}

# Write synthetic data file
synthetic_file = os.path.join(tasks_dir, 'R2R_literal_speaker_data_augmentation_paths.json')
with open(synthetic_file, 'w') as f:
    json.dump([mock_data], f)

# Write train data file
train_file = os.path.join(tasks_dir, 'R2R_train.json')
with open(train_file, 'w') as f:
    json.dump([mock_data], f)

# Write val data file
val_file = os.path.join(tasks_dir, 'R2R_val.json')
with open(val_file, 'w') as f:
    json.dump([mock_data], f)

# Change to temp directory so relative paths work
original_cwd = os.getcwd()
os.chdir(temp_dir)

ARGS = {
    'split': 'train'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
load_datasets Ran
ran on a synthesized input
cacosandon/are-you-looking/tasks/R2R-pano/utils.py
code served (permissive licence) · get_code("247ba71d3fffa065")
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 json
import tempfile
import os

# Create temporary directory and mock JSON files
tmpdir = tempfile.mkdtemp()
instructions_dir = os.path.join(tmpdir, 'instructions')
os.makedirs(instructions_dir, exist_ok=True)

# Create mock data files
train_data = [{"instruction": "go forward", "path": [0, 1, 2]}]
val_seen_data = [{"instruction": "turn left", "path": [1, 2, 3]}]
val_unseen_data = [{"instruction": "turn right", "path": [2, 3, 4]}]
craft_seen_data = [{"instruction": "stop", "path": [3, 4, 5]}]
craft_unseen_data = [{"instruction": "continue", "path": [4, 5, 6]}]
test_data = [{"instruction": "move", "path": [5, 6, 7]}]
synthetic_data = [{"instruction": "synthetic instruction", "path": [6, 7, 8]}]

# Write mock JSON files
with open(os.path.join(instructions_dir, 'R2R_train.json'), 'w') as f:
    json.dump(train_data, f)

with open(os.path.join(instructions_dir, 'R2R_val_seen.json'), 'w') as f:
    json.dump(val_seen_data, f)

with open(os.path.join(instructions_dir, 'R2R_val_unseen.json'), 'w') as f:
    json.dump(val_unseen_data, f)

with open(os.path.join(instructions_dir, 'R2R_craft_seen.json'), 'w') as f:
    json.dump(craft_seen_data, f)

with open(os.path.join(instructions_dir, 'R2R_craft_unseen.json'), 'w') as f:
    json.dump(craft_unseen_data, f)

with open(os.path.join(instructions_dir, 'R2R_test.json'), 'w') as f:
    json.dump(test_data, f)

with open(os.path.join(instructions_dir, 'augmented_data.json'), 'w') as f:
    json.dump(synthetic_data, f)

# Create options object
class Opts:
    def __init__(self):
        self.instructions = instructions_dir
        self.train_instructions = 'train'
        self.path_data_augmentation = 'augmented_data.json'

opts = Opts()

ARGS = {
    'splits': ['train', 'val_seen'],
    'opts': opts
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
load_nav_graph Ran
ran on a synthesized input
cacosandon/are-you-looking/visualization/trajectory_visualization.py
code served (permissive licence) · get_code("9fcdeeb234f31f72")
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 Graph. 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 json
import os
import tempfile
import numpy as np

# Create a temporary directory and connectivity file for testing
temp_dir = tempfile.mkdtemp()
connectivity_dir = os.path.join(temp_dir, 'connectivity')
os.makedirs(connectivity_dir, exist_ok=True)

# Create a sample connectivity JSON file
sample_data = [
    {
        'image_id': 'node_0',
        'included': True,
        'pose': [0, 0, 0, 1, 0, 0, 0, 2, 0, 0, 0, 3],
        'unobstructed': [True, True, False]
    },
    {
        'image_id': 'node_1',
        'included': True,
        'pose': [0, 0, 0, 4, 0, 0, 0, 5, 0, 0, 0, 6],
        'unobstructed': [True, True, False]
    },
    {
        'image_id': 'node_2',
        'included': False,
        'pose': [0, 0, 0, 7, 0, 0, 0, 8, 0, 0, 0, 9],
        'unobstructed': [False, False, True]
    }
]

connectivity_file = os.path.join(connectivity_dir, 'test_connectivity.json')
with open(connectivity_file, 'w') as f:
    json.dump(sample_data, f)

ARGS = {
    'graph_path': connectivity_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Graph — a Graph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Graph output the recorded ndim 1 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
load_nav_graphs Ran
ran; our contract draft was wrong, not the code
CAVED123/regretful-agent/tasks/R2R-pano/utils.py
code served (permissive licence) · get_code("e3a90df1a875a38e")
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 json
import numpy as np
import networkx as nx
from unittest.mock import patch, mock_open
import tempfile
import os

# Create mock connectivity JSON data
mock_connectivity_data = [
    {
        'image_id': 'img_0',
        'included': True,
        'pose': [0, 0, 0, 1.0, 0, 0, 0, 2.0, 0, 0, 0, 3.0],
        'unobstructed': [True, True, False]
    },
    {
        'image_id': 'img_1',
        'included': True,
        'pose': [0, 0, 0, 1.5, 0, 0, 0, 2.5, 0, 0, 0, 3.5],
        'unobstructed': [True, True, False]
    },
    {
        'image_id': 'img_2',
        'included': False,
        'pose': [0, 0, 0, 2.0, 0, 0, 0, 3.0, 0, 0, 0, 4.0],
        'unobstructed': [False, False, True]
    }
]

# Create a temporary directory with mock connectivity files
temp_dir = tempfile.mkdtemp()
connectivity_dir = os.path.join(temp_dir, 'connectivity')
os.makedirs(connectivity_dir, exist_ok=True)

# Write mock connectivity files
for scan_name in ['scan_0', 'scan_1']:
    filepath = os.path.join(connectivity_dir, f'{scan_name}_connectivity.json')
    with open(filepath, 'w') as f:
        json.dump(mock_connectivity_data, f)

# Patch the file open to use our temp directory
original_open = open

def patched_open(file, *args, **kwargs):
    if 'connectivity/' in str(file):
        # Replace connectivity/ path with our temp directory path
        file = file.replace('connectivity/', connectivity_dir + '/')
    return original_open(file, *args, **kwargs)

# Apply the patch
import builtins
builtins.open = patched_open

ARGS = {
    'scans': ['scan_0', 'scan_1']
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx, numpy — 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
proj_masking Ran
ran on a synthesized input
CAVED123/regretful-agent/tasks/R2R-pano/models/modules.py
code served (permissive licence) · get_code("37024446ae219cc6")
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, 5) 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 as nn

torch.manual_seed(0)

# Create a simple projector module that takes input and projects it
class SimpleProjector(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

# Set up dimensions
batch_size = 2
seq_length = 3
input_dim = 4
output_dim = 5

# Create feature tensor: (batch_size, seq_length, input_dim)
feat = torch.randn(batch_size, seq_length, input_dim)

# Create projector
projector = SimpleProjector(input_dim, output_dim)

# Create mask tensor: (batch_size, seq_length)
# This will be unsqueezed to (batch_size, seq_length, 1) and expanded to match proj_feat
mask = torch.ones(batch_size, seq_length)

ARGS = {
    "feat": feat,
    "projector": projector,
    "mask": mask
}
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, 5) — shape (2, 3, 5) 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
load_datasets Not yet run
unverified (harvested, no recorded run)
CAVED123/regretful-agent/tasks/R2R-pano/utils.py
code served (permissive licence) · get_code("be2e80a5268aabb2")
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, 6 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 2 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.

FunctionAlso present inEarliest-datedOthers named
load_nav_graphs 7 other papers
1 dated earlier, 6 later
1901.03035 (2019-01)
Self-Monitoring Navigation Agent via Auxiliary Progress Estimation
1904.04195 (2019-04)
Learning to Navigate Unseen Environments: Back Translation with Environmental Dropout
2010.07954 (2020-10)
Room-Across-Room: Multilingual Vision-and-Language Navigation with Dense Spatiotemporal Grounding
and 4 more not named here
build_vocab 3 other papers
1 dated earlier, 2 later
1901.03035 (2019-01)
Self-Monitoring Navigation Agent via Auxiliary Progress Estimation
1904.04195 (2019-04)
Learning to Navigate Unseen Environments: Back Translation with Environmental Dropout
2107.07201 (2021-07)
Neighbor-view Enhanced Model for Vision and Language Navigation
build_mlp 1 other paper
1 dated earlier
1901.03035 (2019-01)
Self-Monitoring Navigation Agent via Auxiliary Progress Estimation
create_mask 1 other paper
1 dated earlier
1901.03035 (2019-01)
Self-Monitoring Navigation Agent via Auxiliary Progress Estimation
load_datasets 1 other paper
1 dated earlier
1901.03035 (2019-01)
Self-Monitoring Navigation Agent via Auxiliary Progress Estimation
proj_masking 1 other paper
1 dated earlier
1901.03035 (2019-01)
Self-Monitoring Navigation Agent via Auxiliary Progress Estimation

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

As deep learning continues to make progress for challenging perception tasks, there is increased interest in combining vision, language, and decision-making. Specifically, the Vision and Language Navigation (VLN) task involves navigating to a goal purely from language instructions and visual information without explicit knowledge of the goal. Recent successful approaches have made in-roads in achieving good success rates for this task but rely on beam search, which thoroughly explores a large number of trajectories and is unrealistic for applications such as robotics. In this paper, inspired by the intuition of viewing the problem as search on a navigation graph, we propose to use a progress monitor developed in prior work as a learnable heuristic for search. We then propose two modules incorporated into an end-to-end architecture: 1) A learned mechanism to perform backtracking, which decides whether to continue moving forward or roll back to a previous state (Regret Module) and 2) A mechanism to help the agent decide which direction to go next by showing directions that are visited and their associated progress estimate (Progress Marker). Combined, the proposed approach significantly outperforms current state-of-the-art methods using greedy action selection, with 5% absolute improvement on the test server in success rates, and more importantly 8% on success rates normalized by the path length. Our code is available at https://github.com/chihyaoma/regretful-agent .

For agents

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

get_harvested_code_for_paper("1903.01602")
get_code_for_paper("1903.01602")
have("1903.01602")

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

curl https://syntology.ai/api/ran/1903.01602.json

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

[![Syntology run record](https://syntology.ai/api/ran/1903.01602.svg)](https://syntology.ai/paper/1903.01602)

Connect an agent — have() is free.