SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.19740 · 2023

Exploring the Reliability of Large Language Models as Customized Evaluators for Diverse NLP Tasks

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 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 0 have no contract verdict; 4 failures of our instrument; 4 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/2310.19740.json · badge: /api/ran/2310.19740.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
qtli/coeval canonical (classifier label)
classifier basis: census
5 ran · 4 no recorded run
FunctionStatusWhere it lives
apply_rotary_pos_emb Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
qtli/coeval/webapp/monkey_patch_non_inplace.py
code served (permissive licence) · get_code("bdd3c4ef3723890a")
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 5, 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)

# Analyze the function:
# - q, k: query and key tensors, used in element-wise operations with cos/sin
# - cos, sin: gathered from based on position_ids, shape must support gather at dim 2
# - position_ids: indices tensor, shape (batch, seq_len)
# 
# The function:
# 1. Expands position_ids to shape (batch, 1, seq_len, 1)
# 2. Repeats to (batch, num_heads, seq_len, 1)
# 3. Gathers cos/sin at dimension 2 using these indices
# 4. cos and sin are repeated to match batch size
# 5. q and k are multiplied element-wise with cos/sin results
#
# For gathering to work:
# - cos/sin must have at least 3 dimensions with dim 2 being the position dimension
# - cos/sin shape before repeat: (1, num_heads, max_positions, head_dim)
# - After repeat for batch: (batch, num_heads, max_positions, head_dim)
#
# q and k shapes must match cos/sin shapes after gathering

batch_size = 2
num_heads = 3
seq_len = 4
head_dim = 2
max_positions = 5

position_ids = torch.arange(seq_len).unsqueeze(0).repeat(batch_size, 1)
# position_ids shape: (batch_size, seq_len) = (2, 4)

q = torch.randn(batch_size, num_heads, seq_len, head_dim)
# q shape: (2, 3, 4, 2)

k = torch.randn(batch_size, num_heads, seq_len, head_dim)
# k shape: (2, 3, 4, 2)

cos = torch.randn(1, num_heads, max_positions, head_dim)
# cos shape: (1, 3, 5, 2)

sin = torch.randn(1, num_heads, max_positions, head_dim)
# sin shape: (1, 3, 5, 2)

ARGS = {
    'q': q,
    'k': k,
    'cos': cos,
    'sin': sin,
    'position_ids': position_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, ndim 5, 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
compress Ran
ran; our contract draft was wrong, not the code
qtli/coeval/webapp/compression.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Git-King-Zhan/LoReC, is pointer only (licence: NONE) · get_code("80dde39da1ba3708")
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 Git-King-Zhan/LoReC, 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.

firstletterupper Ran
ran; our contract draft was wrong, not the code
qtli/coeval/webapp/evaluator.py
code served (permissive licence) · get_code("1660aaf605a919cd")
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

def firstletterupper(text):
    return text[0].upper() + text[1:]

ARGS = {
    "text": "hello world"
}
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
rotate_half Ran
ran; our contract draft was wrong, not the code
qtli/coeval/webapp/monkey_patch_non_inplace.py
code served (permissive licence) · get_code("62a553b3b2dcaca7")
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)

# The function rotate_half takes a single parameter 'x' which should be a tensor
# It accesses x[..., :x.shape[-1] // 2] and x[..., x.shape[-1] // 2:]
# So x needs to be a tensor with a last dimension that is even-sized

# Create a small tensor with batch size 2 and feature dimension 4
x = torch.randn(2, 4)

ARGS = {
    'x': x
}
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
violates_moderation Ran
ran, violated the contract we drafted
qtli/coeval/webapp/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Mi3-Lab/doScenes-VLM-Planning, is pointer only (licence: AGPL-3.0) · get_code("f9939a84b9a65279")
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 Mi3-Lab/doScenes-VLM-Planning, and AGPL-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.

build_logger Not yet run
unverified (harvested, no recorded run)
qtli/coeval/webapp/utils.py
code served (permissive licence) · get_code("2cd3e6b1d3d305ca")
How it ran

no invocation recorded.

decompress Not yet run
unverified (harvested, no recorded run)
qtli/coeval/webapp/compression.py
code served (permissive licence) · get_code("a2b79de775636d17")
How it ran

no invocation recorded.

forward Not yet run
unverified (harvested, no recorded run)
qtli/coeval/webapp/monkey_patch_non_inplace.py
code served (permissive licence) · get_code("2cbe6ed055d5fd44")
How it ran

no invocation recorded.

pretty_print_semaphore Not yet run
unverified (harvested, no recorded run)
qtli/coeval/webapp/utils.py
code served (permissive licence) · get_code("37899f22fb191b37")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 7 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 4 the same month, 6 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
violates_moderation 56 other papers
5 dated earlier, 2 the same month, 49 later
2106.06909 (2021-06)
GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio
2308.03279 (2023-08)
UniversalNER: TARGETED DISTILLATION FROM LARGE LANGUAGE MODELS FOR OPEN NAMED ENTITY RECOGNITION
2308.13437 (2023-08)
Position-Enhanced Visual Instruction Tuning for Multimodal Large Language Models
and 53 more not named here
pretty_print_semaphore 54 other papers
4 dated earlier, 2 the same month, 48 later
2106.06909 (2021-06)
GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10,000 Hours of Transcribed Audio
2308.13437 (2023-08)
Position-Enhanced Visual Instruction Tuning for Multimodal Large Language Models
2309.07120 (2023-09)
Sight Beyond Text: Multi-Modal Training Enhances LLMs in Truthfulness and Ethics
and 51 more not named here
apply_rotary_pos_emb 47 other papers
13 dated earlier, 4 the same month, 30 later
2305.14167 (2023-05)
DetGPT: Detect What You Need via Reasoning
2305.16103 (2023-05)
ChatBridge: Bridging Modalities with Large Language Model as a Language Catalyst
2305.16355 (2023-05)
PandaGPT: One Model To Instruction-Follow Them All
and 44 more not named here
compress 17 other papers
4 dated earlier, 2 the same month, 11 later
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2303.06865 (2023-03)
FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU
2304.10453 (2023-04)
Phoenix: Democratizing ChatGPT across Languages
and 14 more not named here
decompress 3 other papers
3 dated earlier
2303.06865 (2023-03)
FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU
2304.10453 (2023-04)
Phoenix: Democratizing ChatGPT across Languages
2307.12981 (2023-07)
3D-LLM: Injecting the 3D World into Large Language Models
forward 1 other paper
1 later
2403.00813 (2024-03)
UrbanGPT: Spatio-Temporal Large Language Models
rotate_half 1 other paper
1 later
2403.00813 (2024-03)
UrbanGPT: Spatio-Temporal Large Language Models

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

Previous work adopts large language models (LLMs) as evaluators to evaluate natural language process (NLP) tasks. However, certain shortcomings, e.g., fairness, scope, and accuracy, persist for current LLM evaluators. To analyze whether LLMs can serve as reliable alternatives to humans, we examine the fine-grained alignment between LLM evaluators and human annotators, particularly in understanding the target evaluation tasks and conducting evaluations that meet diverse criteria. This paper explores both conventional tasks (e.g., story generation) and alignment tasks (e.g., math reasoning), each with different evaluation criteria. Our analysis shows that 1) LLM evaluators can generate unnecessary criteria or omit crucial criteria, resulting in a slight deviation from the experts. 2) LLM evaluators excel in general criteria, such as fluency, but face challenges with complex criteria, such as numerical reasoning. We also find that LLM-pre-drafting before human evaluation can help reduce the impact of human subjectivity and minimize annotation outliers in pure human evaluation, leading to more objective evaluation.

For agents

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

get_harvested_code_for_paper("2310.19740")
get_code_for_paper("2310.19740")
have("2310.19740")

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

curl https://syntology.ai/api/ran/2310.19740.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.19740.svg)](https://syntology.ai/paper/2310.19740)

Connect an agent — have() is free.