SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.11976 · NeurIPS · 2021

One Question Answering Model for Many Languages with Cross-lingual Dense Passage Retrieval

Jungo Kasai, Hannaneh Hajishirzi, Akari Asai, Xinyan Yu

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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 0 have no contract verdict; 2 failures of our instrument; 0 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/2107.11976.json · badge: /api/ran/2107.11976.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
AkariAsai/CORA canonical (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
evaluate_batch_e2e Ran
ran; our contract draft was wrong, not the code
AkariAsai/CORA/mGEN/eval_mgen.py
code served (permissive licence) · get_code("0d93b579cc96d791")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import logging

torch.manual_seed(0)

logging.basicConfig(level=logging.INFO)

class SimpleTokenizer:
    def batch_encode_plus(self, texts, return_tensors=None, padding=False, truncation=False):
        # Return a simple dict with input_ids and attention_mask
        batch_size = len(texts)
        seq_len = 5
        input_ids = torch.randint(0, 100, (batch_size, seq_len))
        attention_mask = torch.ones(batch_size, seq_len)
        
        class EncodedOutput:
            pass
        
        output = EncodedOutput()
        output.input_ids = input_ids
        output.attention_mask = attention_mask
        return output
    
    def batch_decode(self, sequences, skip_special_tokens=False):
        # Return dummy decoded strings
        if isinstance(sequences, dict):
            sequences = sequences.get('sequences', sequences)
        batch_size = sequences.shape[0] if hasattr(sequences, 'shape') else len(sequences)
        return [f"Answer {i}" for i in range(batch_size)]


class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 4)
    
    def generate(self, input_ids, attention_mask=None, num_beams=1, min_length=1, 
                 max_length=20, early_stopping=False, num_return_sequences=1, 
                 bad_words_ids=None, output_scores=False, return_dict_in_generate=False):
        batch_size = input_ids.shape[0]
        seq_len = 5
        
        sequences = torch.randint(0, 100, (batch_size, seq_len))
        
        if output_scores and return_dict_in_generate:
            sequences_scores = torch.randn(batch_size)
            return {
                'sequences': sequences,
                'sequences_scores': sequences_scores
            }
        else:
            return sequences


class Args:
    def __init__(self):
        self.device = 'cpu'
        self.num_beams = 1
        self.min_length = 1
        self.max_length = 20
        self.output_scores = False
        self.print_predictions = False


ARGS = {
    'args': Args(),
    'model': SimpleModel(),
    'tokenizer': SimpleTokenizer(),
    'questions': ["What is the capital of France?", "What is 2+2?"]
}
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 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_passages Ran
ran; our contract draft was wrong, not the code
AkariAsai/CORA/mDPR/dense_retriever.py
code served (permissive licence) · get_code("161e67ea95835704")
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 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 tempfile
import csv
import os

# Create a temporary TSV file with test data
temp_file = tempfile.NamedTemporaryFile(mode='w', suffix='.tsv', delete=False)
temp_file.write('id\ttitle\ttext\n')
temp_file.write('doc1\tFirst Document\tThis is the first passage\n')
temp_file.write('doc2\tSecond Document\tThis is the second passage\n')
temp_file.write('doc3\tThird Document\tThis is the third passage\n')
temp_file.close()

ctx_file_path = temp_file.name

ARGS = {
    'ctx_file': ctx_file_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 0 later) and 0 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
evaluate_batch_e2e 1 other paper
1 dated earlier
2010.11856 (2020-10)
XOR QA: Cross-lingual Open-Retrieval Question Answering
load_passages 1 other paper
1 dated earlier
2010.11856 (2020-10)
XOR QA: Cross-lingual Open-Retrieval Question Answering

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

We present Cross-lingual Open-Retrieval Answer Generation (CORA), the first unified many-to-many question answering (QA) model that can answer questions across many languages, even for ones without language-specific annotated data or knowledge sources. We introduce a new dense passage retrieval algorithm that is trained to retrieve documents across languages for a question. Combined with a multilingual autoregressive generation model, CORA answers directly in the target language without any translation or in-language retrieval modules as used in prior work. We propose an iterative training method that automatically extends annotated data available only in high-resource languages to low-resource ones. Our results show that CORA substantially outperforms the previous state of the art on multilingual open QA benchmarks across 26 languages, 9 of which are unseen during training. Our analyses show the significance of cross-lingual retrieval and generation in many languages, particularly under low-resource settings. Our code and trained model are publicly available at https://github.com/AkariAsai/ CORA.

For agents

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

get_harvested_code_for_paper("2107.11976")
get_code_for_paper("2107.11976")
have("2107.11976")

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

curl https://syntology.ai/api/ran/2107.11976.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.11976.svg)](https://syntology.ai/paper/2107.11976)

Connect an agent — have() is free.