SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.13837 · ACL · 2020

Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs

Sung Hwang, Dong Lee, South Korea, Seanie Lee, Donghwan Kim, Woo Jeong

arXiv · PDF · Open in the Atlas

Code that ran

11 of 16 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 10 have no contract verdict; 1 failure of our instrument; 5 no recorded run), measured 2026-11-22; 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/2005.13837.json · badge: /api/ran/2005.13837.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
seanie12/Info-HCVAE canonical (classifier label)
classifier basis: census
11 ran · 5 no recorded run
FunctionStatusWhere it lives
batch_to_device Ran
ran on a synthesized input
seanie12/Info-HCVAE/vae/utils.py
code served (permissive licence) · get_code("7cef89edf29127bc")
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)

# Create small test tensors for the batch
# batch is expected to be a tuple/iterable of 5 tensors:
# (c_ids, q_ids, a_ids, start_positions, end_positions)

batch_size = 2
c_max_len = 8
q_max_len = 6

# Context IDs: shape (batch_size, c_max_len) - contains token IDs
c_ids = torch.randint(1, 100, (batch_size, c_max_len))

# Question IDs: shape (batch_size, q_max_len) - contains token IDs
q_ids = torch.randint(1, 100, (batch_size, q_max_len))

# Answer IDs: shape (batch_size, c_max_len) - contains token IDs
a_ids = torch.randint(1, 100, (batch_size, c_max_len))

# Start positions: shape (batch_size,) - scalar positions
start_positions = torch.randint(0, c_max_len, (batch_size,))

# End positions: shape (batch_size,) - scalar positions
end_positions = torch.randint(0, c_max_len, (batch_size,))

# Create batch as a tuple (which is iterable)
batch = (c_ids, q_ids, a_ids, start_positions, end_positions)

# Device to move to
device = 'cpu'

ARGS = {
    'batch': batch,
    'device': device
}
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
cal_attn Ran
ran on a synthesized input
seanie12/Info-HCVAE/vae/models.py
code served (permissive licence) · get_code("1f734017a6f68175")
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
import torch.nn.functional as F

torch.manual_seed(0)

# Build arguments for cal_attn(query, memories, mask)
# From the function body:
# - query is matmul'd with memories.transpose(-1, -2), so query shape should be (batch, ..., hidden_dim)
# - memories is transposed and matmul'd, so memories shape should be (batch, seq_len, hidden_dim)
# - mask is converted to float, subtracted from 1, multiplied by -10000, then added to attn_logits
# - mask should have compatible shape for broadcasting with attn_logits

batch_size = 2
seq_len = 3
hidden_dim = 4

# query: (batch, hidden_dim) - represents a single query per batch item
query = torch.randn(batch_size, hidden_dim)

# memories: (batch, seq_len, hidden_dim) - represents sequence of memory vectors
memories = torch.randn(batch_size, seq_len, hidden_dim)

# mask: (batch, seq_len) - binary mask where 0 means attend, 1 means mask out
# After (1.0 - mask.float()) * -10000.0, masked positions get large negative values
mask = torch.zeros(batch_size, seq_len)
mask[0, 2] = 1.0  # mask out last position in first batch item

ARGS = {
    "query": query,
    "memories": memories,
    "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 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
convert_examples_to_features_answer_id Ran
ran on a synthesized input
seanie12/Info-HCVAE/qa-eval/squad_utils.py
code served (permissive licence) · get_code("1e2aae97388cfe6e")
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 collections
from copy import deepcopy

# Define a minimal tokenizer that supports the required methods
class SimpleTokenizer:
    def tokenize(self, text):
        # Simple whitespace tokenization
        return text.lower().split()
    
    def convert_tokens_to_ids(self, tokens):
        # Map tokens to simple integer ids
        return [hash(token) % 1000 for token in tokens]

# Define a minimal Example class with required attributes
class Example:
    def __init__(self, question_text, doc_tokens, start_position, end_position, 
                 orig_answer_text, is_impossible):
        self.question_text = question_text
        self.doc_tokens = doc_tokens
        self.start_position = start_position
        self.end_position = end_position
        self.orig_answer_text = orig_answer_text
        self.is_impossible = is_impossible

# Create minimal examples for testing
examples = [
    Example(
        question_text="What is the capital?",
        doc_tokens=["The", "capital", "is", "Paris", "in", "France"],
        start_position=2,
        end_position=3,
        orig_answer_text="Paris",
        is_impossible=False
    ),
    Example(
        question_text="Where is it located?",
        doc_tokens=["France", "is", "located", "in", "Europe"],
        start_position=0,
        end_position=0,
        orig_answer_text="France",
        is_impossible=False
    )
]

tokenizer = SimpleTokenizer()

ARGS = {
    'examples': examples,
    'tokenizer': tokenizer,
    'max_seq_length': 128,
    'doc_stride': 64,
    'max_query_length': 32,
    'max_ans_length': 16,
    'is_training': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm — 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
get_linear_schedule_with_warmup Ran
ran; our contract draft was wrong, not the code
seanie12/Info-HCVAE/qa-eval/trainer.py
code served (permissive licence) · get_code("a3c458ae06a14711")
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 LambdaLR. 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 torch.optim import SGD

torch.manual_seed(0)

# Create a simple model
model = nn.Linear(4, 3)

# Create an optimizer
optimizer = SGD(model.parameters(), lr=0.1)

# Define arguments for get_linear_schedule_with_warmup
ARGS = {
    "optimizer": optimizer,
    "num_warmup_steps": 100,
    "num_training_steps": 1000,
    "last_epoch": -1
}
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 LambdaLR — a LambdaLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LambdaLR 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
gumbel_softmax Ran
ran on a synthesized input
seanie12/Info-HCVAE/vae/models.py
code served (permissive licence) · get_code("7b8c9c000eba3141")
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

torch.manual_seed(0)

# Create small logits tensor (batch=2, classes=4)
logits = torch.randn(2, 4)

# Define all parameters
ARGS = {
    'logits': logits,
    'tau': 1.0,
    'hard': False,
    'eps': 1e-20,
    'dim': -1
}
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
post_process Ran
ran on a synthesized input
seanie12/Info-HCVAE/vae/translate.py
code served (permissive licence) · get_code("845c524066f993b4")
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)

# Create small tensors for the function
batch_size = 2
max_len = 10
total_max_len = 384

# q_ids: batch_size x seq_len, containing token IDs (non-zero values) and padding (zeros)
q_ids = torch.tensor([
    [101, 2054, 2003, 0, 0, 0, 0, 0, 0, 0],  # [CLS] What is [PAD]...
    [101, 2178, 0, 0, 0, 0, 0, 0, 0, 0]      # [CLS] Who [PAD]...
], dtype=torch.long)

# c_ids: batch_size x seq_len, context IDs starting with [SEP]
c_ids = torch.tensor([
    [102, 2054, 2003, 2009, 0, 0, 0, 0, 0, 0],  # [SEP] What is it [PAD]...
    [102, 2178, 2003, 0, 0, 0, 0, 0, 0, 0]      # [SEP] Who is [PAD]...
], dtype=torch.long)

# start_positions: batch_size, positions in context (adjusted to be > 0)
start_positions = torch.tensor([2, 1], dtype=torch.long)

# end_positions: batch_size, positions in context (adjusted to be > 0)
end_positions = torch.tensor([3, 2], dtype=torch.long)

ARGS = {
    "q_ids": q_ids,
    "start_positions": start_positions,
    "end_positions": end_positions,
    "c_ids": c_ids,
    "total_max_len": 384
}
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
progress_bar Ran
ran on a synthesized input
seanie12/Info-HCVAE/qa-eval/utils.py
code served (permissive licence) · get_code("58fb608bcfbbdffe")
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 str. 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

ARGS = {
    "completed": 75,
    "total": 100,
    "step": 5
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
return_mask_lengths Ran
ran on a synthesized input
seanie12/Info-HCVAE/vae/models.py
code served (permissive licence) · get_code("b058372cf5b99e2b")
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)

# The function expects 'ids' which is a tensor
# It applies torch.sign() and torch.sum() on dimension 1
# So ids should be a 2D tensor with shape (batch_size, seq_length)
# Using batch_size=2, seq_length=4 for small but distinct dimensions

ids = torch.tensor([
    [1, 2, 0, 3],
    [5, 0, 0, 2]
], dtype=torch.float32)

ARGS = {
    "ids": ids
}
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
return_mask_lengths Ran
ran on a synthesized input
seanie12/Info-HCVAE/vae/translate.py
code served (permissive licence) · get_code("c058fcfe8ca35ccf")
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)

# The function expects ids as a tensor
# It uses torch.sign(ids) to create a mask, then sums along dimension 1
# So ids should be a 2D tensor with shape (batch_size, sequence_length)
# Using small distinct dimensions: batch=2, sequence=4

ids = torch.tensor([
    [1, 2, 0, 3],
    [0, 5, 2, 0]
], dtype=torch.float32)

ARGS = {
    "ids": ids
}
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
time_since Ran
ran on a synthesized input
seanie12/Info-HCVAE/qa-eval/utils.py
code served (permissive licence) · get_code("3d69d712283cfd10")
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 float, 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 time

# The function time_since takes a single parameter 't' which should be a float
# representing a previous timestamp. The function computes the elapsed time
# by subtracting this value from the current time.

# Create a timestamp from slightly in the past
t = time.time() - 0.5  # 0.5 seconds ago

ARGS = {
    "t": t
}
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 float, 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
user_friendly_time Ran
ran on a synthesized input
seanie12/Info-HCVAE/qa-eval/utils.py
code served (permissive licence) · get_code("0a7669f48a1dc3a1")
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 str. 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 = {
    's': 3661
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
convert_examples_to_features Not yet run
unverified (harvested, no recorded run)
seanie12/Info-HCVAE/qa-eval/squad_utils.py
code served (permissive licence) · get_code("72c5000d831f12e0")
How it ran

no invocation recorded.

convert_examples_to_harv_features Not yet run
unverified (harvested, no recorded run)
seanie12/Info-HCVAE/qa-eval/squad_utils.py
code served (permissive licence) · get_code("4481bd2f1ff50f5b")
How it ran

no invocation recorded.

cook_refs Not yet run
unverified (harvested, no recorded run)
seanie12/Info-HCVAE/vae/qgevalcap/cider/cider_scorer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from vrama91/coco-caption, is pointer only (licence: NOASSERTION) · get_code("5172aba957a32d9e")
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 vrama91/coco-caption, 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.

cook_test Not yet run
unverified (harvested, no recorded run)
seanie12/Info-HCVAE/vae/qgevalcap/cider/cider_scorer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from vrama91/coco-caption, is pointer only (licence: NOASSERTION) · get_code("90a4bcbf05586ff3")
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 vrama91/coco-caption, 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.

precook Not yet run
unverified (harvested, no recorded run)
seanie12/Info-HCVAE/vae/qgevalcap/cider/cider_scorer.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from vrama91/coco-caption, is pointer only (licence: NOASSERTION) · get_code("422c41f5b22cf9c3")
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 vrama91/coco-caption, 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 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 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, 3 later) and 12 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
cook_refs 25 other papers
12 dated earlier, 13 later
1411.5726 (2014-11)
CIDEr: Consensus-based Image Description Evaluation
1502.03044 (2015-02)
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
1503.01817 (2015-03)
YFCC100M: The New Data in Multimedia Research
and 22 more not named here
cook_test 22 other papers
9 dated earlier, 13 later
1411.5726 (2014-11)
CIDEr: Consensus-based Image Description Evaluation
1502.03044 (2015-02)
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
1503.01817 (2015-03)
YFCC100M: The New Data in Multimedia Research
and 19 more not named here
precook 8 other papers
8 dated earlier
1411.5726 (2014-11)
CIDEr: Consensus-based Image Description Evaluation
1502.03044 (2015-02)
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
1503.01817 (2015-03)
YFCC100M: The New Data in Multimedia Research
and 5 more not named here
get_linear_schedule_with_warmup 6 other papers
6 later
2106.11310 (2021-06)
Towards Long-Form Video Understanding
2106.13353 (2021-06)
Cutting Down on Prompts and Parameters: Simple Few-Shot Learning with Language Models
2204.07275 (2022-04)
Incremental Prompting: Episodic Memory Prompt for Lifelong Event Detection
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

One of the most crucial challenges in question answering (QA) is the scarcity of labeled data, since it is costly to obtain question-answer (QA) pairs for a target text domain with human annotation. An alternative approach to tackle the problem is to use automatically generated QA pairs from either the problem context or from large amount of unstructured texts (e.g. Wikipedia). In this work, we propose a hierarchical conditional variational autoencoder (HCVAE) for generating QA pairs given unstructured texts as contexts, while maximizing the mutual information between generated QA pairs to ensure their consistency. We validate our Information Maximizing Hierarchical Conditional Variational AutoEncoder (Info-HCVAE) on several benchmark datasets by evaluating the performance of the QA model (BERT-base) using only the generated QA pairs (QA-based evaluation) or by using both the generated and human-labeled pairs (semisupervised learning) for training, against stateof-the-art baseline models. The results show that our model obtains impressive performance gains over all baselines on both tasks, using only a fraction of data for training. 1 * Equal contribution 1 The generated QA pairs and the code can be found at https://github.com/seanie12/Info-HCVAE

For agents

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

get_harvested_code_for_paper("2005.13837")
get_code_for_paper("2005.13837")
have("2005.13837")

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

curl https://syntology.ai/api/ran/2005.13837.json

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

[![Syntology run record](https://syntology.ai/api/ran/2005.13837.svg)](https://syntology.ai/paper/2005.13837)

Connect an agent — have() is free.