SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.06175 · ACL · 2022

Efficient Cluster-Based k-Nearest-Neighbor Machine Translation

Boxing Chen, Kai Fan, Deyi Xiong, Dexin Wang

arXiv · PDF · Open in the Atlas

Code that ran

4 of 6 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 4 have no contract verdict; 0 failures of our instrument; 2 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/2204.06175.json · badge: /api/ran/2204.06175.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
cordercorder/knn-models community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
FunctionStatusWhere it lives
build_flatl2_index Ran
ran on a synthesized input
cordercorder/knn-models/knn_models_cli/build_faiss_index.py
code served (permissive licence) · get_code("a3b087d2f3431832")
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 IndexFlatL2. 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 numpy as np
import faiss
import logging

logging.basicConfig(level=logging.INFO)

# Create datastore keys: shape (num_vectors, embedding_dim)
# Using distinct dimensions: 5 vectors of dimension 4
datastore_keys = np.random.randn(5, 4).astype(np.float32)

ARGS = {
    'datastore_keys': datastore_keys,
    'use_gpu': False,
    'knn_fp16': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
faiss, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type IndexFlatL2 — a IndexFlatL2 output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a IndexFlatL2 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_retrieved_text_to_tensor Ran
ran on a synthesized input
cordercorder/knn-models/knn_models/es_knn_utils.py
code served (permissive licence) · get_code("e86685036268a1d5")
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 2, 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
import numpy as np

torch.manual_seed(0)

class DummyDictionary:
    """Stand-in dictionary with encode_line method"""
    def __init__(self):
        self.vocab = {"hello": 0, "world": 1, "test": 2, "data": 3}
    
    def encode_line(self, line, add_if_not_exist=False):
        """Encode a line of text to token IDs"""
        tokens = [self.vocab.get(word, 3) for word in line.split()]
        return torch.tensor(tokens, dtype=torch.long)

# Create dummy dictionary
dictionary = DummyDictionary()

# Create retrieved_text: list of lists of strings
retrieved_text = [
    ["hello world", "test data"],
    ["hello test", "world data"]
]

# Create retrieved_text_ids: list of lists of IDs (matching structure)
retrieved_text_ids = [
    [1, 2],
    [3, 4]
]

ARGS = {
    "retrieved_text": retrieved_text,
    "retrieved_text_ids": retrieved_text_ids,
    "dictionary": dictionary
}
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 2, dtype family int — 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
edit_distance Ran
ran on a synthesized input
cordercorder/knn-models/knn_models/es_knn_utils.py
code served (permissive licence) · get_code("c9f0e2e7c0c48ee3")
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 torch

torch.manual_seed(0)

# edit_distance takes two string parameters
string1 = "kitten"
string2 = "sitting"

ARGS = {
    "string1": string1,
    "string2": string2,
}
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 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
get_captured_module Ran
ran on a synthesized input
cordercorder/knn-models/knn_models/knn_utils.py
code served (permissive licence) · get_code("69d1f014163532b4")
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 InnerModule. 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
import logging

# Set up logging
logging.basicConfig(level=logging.INFO)

# Create a nested module structure that the function will traverse
class InnerModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

class MiddleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.layers = nn.ModuleList([
            InnerModule(),
            InnerModule(),
            InnerModule()
        ])
    
    def forward(self, x):
        return x

class DecoderModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.encoder = MiddleModule()
        self.decoder_head = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.decoder_head(x)

torch.manual_seed(0)

# Construct the decoder
decoder = DecoderModule()

# Test case 1: Simple attribute access
captured_module_name_1 = "encoder"

# Test case 2: Nested attribute access with indexing
captured_module_name_2 = "encoder.layers[-1]"

# We'll use the second test case as it exercises more of the function
ARGS = {
    "decoder": decoder,
    "captured_module_name": "encoder.layers[-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 InnerModule — a InnerModule output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a InnerModule 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
build_ivfpq_index Not yet run
unverified (harvested, no recorded run)
cordercorder/knn-models/knn_models_cli/build_faiss_index.py
code served (permissive licence) · get_code("72cd21cbd7c41fad")
How it ran

no invocation recorded.

get_gpu_resource Not yet run
unverified (harvested, no recorded run)
cordercorder/knn-models/knn_models_cli/build_faiss_index.py
code served (permissive licence) · get_code("8ed29d811d387a63")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

k-Nearest-Neighbor Machine Translation (kNN-MT) has been recently proposed as a non-parametric solution for domain adaptation in neural machine translation (NMT). It aims to alleviate the performance degradation of advanced MT systems in translating out-ofdomain sentences by coordinating with an additional token-level feature-based retrieval module constructed from in-domain data. Previous studies (Khandelwal et al., 2021; Zheng et al., 2021a) have already demonstrated that non-parametric NMT is even superior to models fine-tuned on out-of-domain data. In spite of this success, kNN retrieval is at the expense of high latency, in particular for large datastores. To make it practical, in this paper, we explore a more efficient kNN-MT and propose to use clustering to improve the retrieval efficiency. Concretely, we first propose a cluster-based Compact Network for feature reduction in a contrastive learning manner to compress context features into 90+% lower dimensional vectors. We then suggest a cluster-based pruning solution to filter out 10%~40% redundant nodes in large datastores while retaining translation quality. Our proposed methods achieve better or comparable performance while reducing up to 57% inference latency against the advanced non-parametric MT model on several machine translation benchmarks. Experimental results indicate that the proposed methods maintain the most useful information of the original datastore and the Compact Network shows good generalization on unseen domains. Codes are available at https: //github.com/tjunlp-lab/PCKMT.

For agents

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

get_harvested_code_for_paper("2204.06175")
get_code_for_paper("2204.06175")
have("2204.06175")

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

curl https://syntology.ai/api/ran/2204.06175.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.06175.svg)](https://syntology.ai/paper/2204.06175)

Connect an agent — have() is free.