SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.00368 · 2024

Improving Text Embeddings with Large Language Models

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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, 1 violated it and 1 has no contract verdict; 3 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/2401.00368.json · badge: /api/ran/2401.00368.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
kamalkraj/e5-mistral-7b-instruct community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
mcgill-nlp/llm2vec reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
FunctionStatusWhere it lives
append_instruction Ran
ran; our contract draft was wrong, not the code
mcgill-nlp/llm2vec/examples/classification.py
code served (permissive licence) · get_code("1ed15c587eee355c")
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 2, 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

torch.manual_seed(0)

instruction = 'Classify a given Amazon customer review text as either counterfactual or notcounterfactual: '

sentences = [
    "This product is amazing and works great.",
    "The quality is poor and I am disappointed."
]

ARGS = {
    'instruction': instruction,
    'sentences': sentences
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
batch_to_device Ran
ran; our contract draft was wrong, not the code
mcgill-nlp/llm2vec/llm2vec/llm2vec.py
code served (permissive licence) · get_code("74b280cb6a91117e")
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 torch
from torch import Tensor, device, nn

torch.manual_seed(0)

# Create a sample batch dictionary with mixed content
batch = {
    "input_ids": torch.randint(0, 100, (2, 4)),  # batch_size=2, seq_len=4
    "attention_mask": torch.ones(2, 4),
    "token_type_ids": torch.zeros(2, 4),
    "labels": torch.randint(0, 10, (2,)),
    "metadata": "some_string",  # non-tensor value
    "count": 42,  # non-tensor value
}

# Target device (CPU only)
target_device = device("cpu")

ARGS = {
    "batch": batch,
    "target_device": target_device,
}
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 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
get_detailed_instruct Ran
ran on a synthesized input
kamalkraj/e5-mistral-7b-instruct/peft_lora_embedding_semantic_search.py
code served (permissive licence) · get_code("06b718c32af96de4")
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)

task_description = "Find the main topic of the given text"
queries = ["What is machine learning?", "Explain neural networks", "Define deep learning"]

ARGS = {
    "task_description": task_description,
    "queries": queries
}
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
info_nce Ran
ran; our contract draft was wrong, not the code
kamalkraj/e5-mistral-7b-instruct/loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from hku-medai/cate, is pointer only (licence: NONE) · get_code("54b175c46812e410")
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 hku-medai/cate, and GitHub detected no licence in that repository when it was checked. 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.

transpose Ran
ran, violated the contract we drafted
kamalkraj/e5-mistral-7b-instruct/loss.py
code served (permissive licence) · get_code("55f1419d4d8c483c")
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 Tensor, ndim 4, dtype family float, shape (2, 4, 5, 3) 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)

# The transpose function takes a single parameter 'x' and calls .transpose(-2, -1) on it
# This requires x to be a tensor with at least 2 dimensions
# We'll create a small tensor with distinct dimensions to make shape probing unambiguous

x = torch.randn(2, 4, 3, 5)  # 4D tensor with distinct sizes

ARGS = {
    "x": x
}
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 Tensor, ndim 4, dtype family float, shape (2, 4, 5, 3) — shape (2, 4, 5, 3) 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 later) and 2 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
transpose 17 other papers
7 dated earlier, 9 later, 1 same year, order not determinable · 2 only as an assembly brick
1910.12037 (2019-10)
Region Mutual Information Loss for Semantic Segmentation
2207.05500 (2022-07)
Modality-Aware Contrastive Instance Learning with Self-Distillation for Weakly-Supervised Audio-Visual Violence Detection
2306.10008 (2023-06)
CLIP2Protect: Protecting Facial Privacy using Text-Guided Makeup via Adversarial Latent Search
and 14 more not named here
info_nce 11 other papers
6 dated earlier, 5 later · 1 only as an assembly brick
2207.05500 (2022-07)
Modality-Aware Contrastive Instance Learning with Self-Distillation for Weakly-Supervised Audio-Visual Violence Detection
2306.10008 (2023-06)
CLIP2Protect: Protecting Facial Privacy using Text-Guided Makeup via Adversarial Latent Search
2307.16377 (2023-07)
JOTR: 3D Joint Contrastive Learning with Transformers for Occluded Human Mesh Recovery
and 8 more not named here
batch_to_device 3 other papers
3 later
2402.09642 (2024-02)
Answer is All You Need: Instruction-following Text Embedding via Answering the Question
2402.15449 (2024-02)
Repetition Improves Language Model Embeddings
2502.12974 (2025-02)
Learning Refined Document Representations for Dense Retrieval via Deliberate Thinking

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

In this paper, we introduce a novel and simple method for obtaining high-quality text embeddings using only synthetic data and less than 1k training steps. Unlike existing methods that often depend on multi-stage intermediate pre-training with billions of weakly-supervised text pairs, followed by fine-tuning with a few labeled datasets, our method does not require building complex training pipelines or relying on manually collected datasets that are often constrained by task diversity and language coverage. We leverage proprietary LLMs to generate diverse synthetic data for hundreds of thousands of text embedding tasks across 93 languages. We then fine-tune open-source decoder-only LLMs on the synthetic data using standard contrastive loss. Experiments demonstrate that our method achieves strong performance on highly competitive text embedding benchmarks without using any labeled data. Furthermore, when fine-tuned with a mixture of synthetic and labeled data, our model sets new state-of-the-art results on the BEIR and MTEB benchmarks.

For agents

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

get_harvested_code_for_paper("2401.00368")
get_code_for_paper("2401.00368")
have("2401.00368")

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

curl https://syntology.ai/api/ran/2401.00368.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.00368.svg)](https://syntology.ai/paper/2401.00368)

Connect an agent — have() is free.