SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.09040 · 2024

RAGGED: Towards Informed Design of Scalable and Stable RAG Systems

arXiv · PDF · Open in the Atlas

Code that ran

13 of 14 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; 5 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 7 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/2403.09040.json · badge: /api/ran/2403.09040.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
neulab/ragged canonical (classifier label)
classifier basis: census
13 ran · 1 no recorded run
FunctionStatusWhere it lives
convert_reader_results_to_zeno Ran
ran; our contract draft was wrong, not the code
neulab/ragged/analysis_framework/compile_results.py
code served (permissive licence) · get_code("f220897243bce737")
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 numpy as np

# Create reader output data with proper structure
reader_output_data = [
    {
        'id': 'q1',
        'input': 'What is the capital of France?',
        'answer': 'Paris',
        'answer_evaluation': {'exact_match': True, 'f1': 1.0},
        'gold_answers': ['Paris', 'paris'],
        'retrieved_passages': [
            {
                'passage_text': 'Paris is the capital of France.',
                'page_id': 'page_1',
                'page_par_id': 'par_1',
                'page_id_match': True,
                'page_par_id_match': True,
                'answer_in_context': True
            },
            {
                'passage_text': 'France is a European country.',
                'page_id': 'page_2',
                'page_par_id': 'par_2',
                'page_id_match': False,
                'page_par_id_match': False,
                'answer_in_context': False
            }
        ]
    },
    {
        'id': 'q2',
        'input': 'Who is the president of the USA?',
        'answer': 'Joe Biden',
        'answer_evaluation': {'exact_match': False, 'f1': 0.5},
        'gold_answers': ['Joe Biden', 'Biden'],
        'retrieved_passages': [
            {
                'passage_text': 'Joe Biden is the current president.',
                'page_id': 'page_3',
                'page_par_id': 'par_3',
                'page_id_match': True,
                'page_par_id_match': True,
                'answer_in_context': True
            }
        ]
    }
]

# Create retriever evaluation data with matching structure
retriever_eval_data = [
    {
        'id': 'q1',
        'passage-level results': [
            {
                'rank': 1,
                'relevance': 0.95
            },
            {
                'rank': 2,
                'relevance': 0.3
            }
        ]
    },
    {
        'id': 'q2',
        'passage-level results': [
            {
                'rank': 1,
                'relevance': 0.88
            }
        ]
    }
]

# Test both baseline and non-baseline cases
ARGS = {
    'reader_output_data': reader_output_data,
    'retriever_eval_data': retriever_eval_data,
    'is_baseline': False
}
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 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_jsonl Ran
ran, honoured the contract we drafted
neulab/ragged/retriever/evaluate_retriever.py
code served (permissive licence) · get_code("ca677f8c2b43120d")
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 int, ndim 0, dtype family int, 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 tempfile
import os

# Create a temporary JSONL file with some test data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.jsonl")

with open(temp_file, 'w') as f:
    f.write('{"key1": "value1"}\n')
    f.write('{"key2": "value2"}\n')
    f.write('{"key3": "value3"}\n')

ARGS = {
    "filename": temp_file
}
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 int, ndim 0, dtype family int, 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
create_prompt Ran
ran; our contract draft was wrong, not the code
neulab/ragged/reader/utils.py
code served (permissive licence) · get_code("289d7efd299b90d4")
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 = {
    'question': 'What is the capital of France?',
    'context': 'France is a country in Western Europe. Its capital city is Paris.'
}
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
get_precision Ran
ran, honoured the contract we drafted
neulab/ragged/analysis_framework/zeno_upload.py
code served (permissive licence) · get_code("1b1fbbd9dac2072c")
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 float64, ndim 0, 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 numpy as np

# Create sample data for the function
# guess_id_set: a set of guessed IDs
# gold_id_set: a set of correct/gold IDs

guess_id_set = {1, 2, 3, 5}
gold_id_set = {1, 2, 4, 6, 7}

ARGS = {
    "guess_id_set": guess_id_set,
    "gold_id_set": gold_id_set,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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_precision Ran
ran, honoured the contract we drafted
neulab/ragged/retriever/evaluate_retriever.py
code served (permissive licence) · get_code("d05429bcacc8eccc")
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 float64, ndim 0, 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 numpy as np

# Analyze the function:
# - get_precision takes two parameters: guess_page_id_set and gold_page_id_set
# - Both are used as sets (membership checking with 'in' operator)
# - The function iterates over guess_page_id_set and checks if each element is in gold_page_id_set
# - Returns the mean of a list of booleans

# Create small example sets with page IDs
guess_page_id_set = {1, 2, 3, 5}
gold_page_id_set = {1, 2, 4, 6}

ARGS = {
    "guess_page_id_set": guess_page_id_set,
    "gold_page_id_set": gold_page_id_set,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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_recall Ran
ran, honoured the contract we drafted
neulab/ragged/analysis_framework/zeno_upload.py
code served (permissive licence) · get_code("19d64b24f5ce24e3")
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 float64, ndim 0, 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 numpy as np

def get_recall(guess_id_set, gold_id_set):
    recall = np.mean([[s in guess_id_set] for s in gold_id_set]) if len(gold_id_set) > 0 else 0.0
    return recall

# Setup: The function expects two sets (or set-like objects) containing IDs
# guess_id_set: a set of guessed/predicted IDs
# gold_id_set: a set of ground truth IDs

ARGS = {
    "guess_id_set": {1, 2, 3, 5},
    "gold_id_set": {1, 2, 4, 6}
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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_recall Ran
ran, honoured the contract we drafted
neulab/ragged/retriever/evaluate_retriever.py
code served (permissive licence) · get_code("a072fc24dd6a64f4")
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 float64, ndim 0, 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 numpy as np

# Build the arguments for get_recall
# The function expects two sets of page IDs

# Create a set of guessed page IDs
guess_page_id_set = {1, 2, 3, 5}

# Create a set of gold (ground truth) page IDs
# Some overlap with guess_page_id_set, some don't
gold_page_id_set = {1, 2, 4, 6}

ARGS = {
    "guess_page_id_set": guess_page_id_set,
    "gold_page_id_set": gold_page_id_set,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
is_potential_number Ran
ran, violated the contract we drafted
neulab/ragged/reader/evaluate_top_k.py
code served (permissive licence) · get_code("507cd94cf077b40b")
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 bool, 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

ARGS = {
    "word": "twenty"
}
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 bool, 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
load_json Ran
ran; our contract draft was wrong, not the code
neulab/ragged/file_utils.py
code served (permissive licence) · get_code("130b876997375207")
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 a temporary JSON file with test data
temp_dir = tempfile.mkdtemp()
test_file = os.path.join(temp_dir, "test_data.json")

test_data = [
    {"id": 3, "name": "Alice", "value": 100},
    {"id": 1, "name": "Bob", "value": 200},
    {"id": 2, "name": "Charlie", "value": 300}
]

with open(test_file, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "filename": test_file,
    "sort_by_id": False
}
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_jsonl Ran
ran; our contract draft was wrong, not the code
neulab/ragged/file_utils.py
code served (permissive licence) · get_code("ca1dbea1a9af7003")
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 a temporary JSONL file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', suffix='.jsonl', delete=False)
temp_filename = temp_file.name

# Write test data to the temporary file
test_data = [
    {'id': 3, 'name': 'Charlie', 'value': 100},
    {'id': 1, 'name': 'Alice', 'value': 50},
    {'id': 2, 'name': 'Bob', 'value': 75},
]

for obj in test_data:
    temp_file.write(json.dumps(obj) + '\n')

temp_file.close()

ARGS = {
    'filename': temp_filename,
    'sort_by_id': True,
}
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
normalize_answer Ran
ran; our contract draft was wrong, not the code
neulab/ragged/reader/evaluate_top_k.py
pointer only (licence: MIT); the stored copy get_code serves, from borealisai/radar-decoding, is pointer only (licence: LGPL-3.0) · get_code("e7e75981cb464788")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from borealisai/radar-decoding, and LGPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

post_process_answers Ran
ran; our contract draft was wrong, not the code
neulab/ragged/reader/utils.py
code served (permissive licence) · get_code("d0896f977a782cd6")
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

torch.manual_seed(0)

# The function post_process_answers takes a list of strings and:
# 1. Strips whitespace from each string
# 2. Splits by newlines and takes the first element
# So we need a list of strings with potential whitespace and newlines

answers = [
    "  Hello\nWorld\n  ",
    "Test answer\nSecond line",
    "\n  Trimmed  \nMultiple\nLines"
]

ARGS = {
    "answers": answers
}
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
truncate_prompt Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
neulab/ragged/reader/utils.py
code served (permissive licence) · get_code("f6e6c7c81db7d3e1")
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 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
from transformers import AutoTokenizer

# Create a minimal stand-in tokenizer that implements the required interface
class StandInTokenizer:
    def __call__(self, text, add_special_tokens=True, max_length=None, truncation=False):
        # Simple tokenization: split by spaces and convert to token ids
        tokens = text.split()
        token_ids = [hash(t) % 10000 for t in tokens]  # Simple hash-based token ids
        
        if max_length and truncation:
            token_ids = token_ids[:max_length]
        
        return {'input_ids': token_ids}
    
    def decode(self, token_ids):
        # Simple decode: convert back to pseudo-words
        words = [f'token_{tid}' for tid in token_ids]
        return ' '.join(words)

# Create tokenizer instance
tokenizer = StandInTokenizer()

# Create sample prompt dict
prompt = {
    'question': 'What is the capital of France?',
    'context': 'France is a country in Western Europe. Its capital is Paris. Paris is known for the Eiffel Tower and many museums.'
}

# Tokenize instruction string to get token count
instruction_str = 'Give simple short one phrase answers for the questions based on the context'
instruction_str_tokens = tokenizer(instruction_str)['input_ids']

# Set parameters
total_tokens = 512
max_new_tokens = 50

ARGS = {
    'prompt': prompt,
    'tokenizer': tokenizer,
    'instruction_str_tokens': instruction_str_tokens,
    'total_tokens': total_tokens,
    'max_new_tokens': max_new_tokens
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — 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 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
get_tokenizer Not yet run
unverified (harvested, no recorded run)
neulab/ragged/utils.py
code served (permissive licence) · get_code("6e7ed506217fac36")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 1 later) and 13 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
normalize_answer 133 other papers
68 dated earlier, 3 the same month, 61 later, 1 undated · 1 only as an assembly brick
1508.06615 (2015-08)
Character-Aware Neural Language Models
1606.05250 (2016-06)
SQuAD: 100,000+ Questions for Machine Comprehension of Text
1608.07905 (2016-08)
Machine Comprehension Using Match-LSTM and Answer Pointer
and 130 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

Retrieval-augmented generation (RAG) enhances language models by integrating external knowledge, but its effectiveness is highly dependent on system configuration. Improper retrieval settings can degrade performance, making RAG less reliable than closed-book generation. In this work, we introduce RAGGED, a framework for systematically evaluating RAG systems across diverse retriever-reader configurations, retrieval depths, and datasets. Our analysis reveals that reader robustness to noise is the key determinant of RAG stability and scalability. Some readers benefit from increased retrieval depth, while others degrade due to their sensitivity to distracting content. Through large-scale experiments on open-domain, multi-hop, and specialized-domain datasets, we show that retrievers, rerankers, and prompts influence performance but do not fundamentally alter these reader-driven trends. By providing a principled framework and new metrics to assess RAG stability and scalability, RAGGED enables systematic evaluation of retrieval-augmented generation systems, guiding future research on optimizing retrieval depth and model robustness.

For agents

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

get_harvested_code_for_paper("2403.09040")
get_code_for_paper("2403.09040")
have("2403.09040")

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

curl https://syntology.ai/api/ran/2403.09040.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.09040.svg)](https://syntology.ai/paper/2403.09040)

Connect an agent — have() is free.