Papers › Can Large Language Models Capture Dissenting Human Voices?

Can Large Language Models Capture Dissenting Human Voices?

23 May 2023arXiv:2305.13788archive 2025-07-28

Noah Lee, Na Min An, James Thorne

Large language models (LLMs) have shown impressive achievements in solving a broad range of tasks. Augmented by instruction fine-tuning, LLMs have also been shown to generalize in zero-shot settings as well. However, whether LLMs closely align with the human disagreement distribution has not been well-studied, especially within the scope of natural language inference (NLI). In this paper, we evaluate the performance and alignment of LLM distribution with humans using two different techniques to estimate the multinomial distribution: Monte Carlo Estimation (MCE) and Log Probability Estimation (LPE). As a result, we show LLMs exhibit limited ability in solving NLI tasks and simultaneously fail to capture human disagreement distribution. The inference and human alignment performances plunge even further on data samples with high human disagreement levels, raising concerns about their natural language understanding (NLU) ability and their representativeness to a larger human population. The source code for the experiments is available at https://github.com/xfactlab/emnlp2023-LLM-Disagreement

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 28 code samples harvested from 2 repositories linked to this paper: 24 ran (of which 0 constructed an object rather than computing a result; 24 with no instrument failure: 0 honoured, 1 violated, 23 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 18 samples from 1 repository, 14 ran (14 with no instrument failure; 0 where Syntology's instrument failed); found in paper text by Syntology: 10 samples from 1 repository, 10 ran (10 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 24 samples that ran has no recorded date.

xfactlab/emnlp2023-llm-disagreement officialmentioned in papermentioned on GitHubMITSyntology: 18 samples, 14 ran (0 instrument-failed) · 4 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

28 samples harvested: 24 ran (of which 0 constructed an object rather than computing a result; 24 with no instrument failure: 0 honoured, 1 violated, 23 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
23ran
4unverified

Licence: 0 of the 28 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

xfactlab/emnlp2023-llm-disagreement official repository 18 samples, 14 ran (0 instrument-failed) · 4 unverified
dict_postprocess sample.py ran MIT (permissive) · 6c55f5bfb0365f70 · report fixture: signature instance:dict/0/?, data_type:str/0/str · returned dict · level recorded 2026-09-26
dict_postprocess utils/processing.py ran MIT (permissive) · 5c81be5eb68c1209 · report fixture: signature dict:dict/0/?, sampled:dict/0/?, data_type:str/0/str, is_logprob:bool/0/bool · returned tuple · level recorded 2026-09-26
entropy_sample sample.py ran MIT (permissive) · a536795ec1c148df · report fixture: signature data:dict/0/?, process_fn:callable/0/?, data_type:str/0/str, sample_size:int/0/int · returned dict · level recorded 2026-09-26
generate_prompt_template utils/prompts.py ran MIT (permissive) · e31192c49a763711 · report fixture: signature premise:str/0/str, hypothesis:str/0/str, options:list/1/str, template_type:int/0/int, idx:int/0/int · returned str · level recorded 2026-09-26
jensen_shannon utils/metrics.py ran fingerprinted MIT (permissive) · 40756590747738d4 · report fixture: signature p:ndarray/1/float, q:ndarray/1/float · returned float · level recorded 2026-09-26
kl_divergence in 86 papers
census: 18 implementations from 18 papers, 15 distinct outputs across 7 buckets, one shared input per bucket
utils/metrics.py ran fingerprinted MIT (permissive) · cfd7ee739cccf02d · report fixture: signature p:ndarray/1/float, q:ndarray/1/float, is_zero:bool/0/bool · returned float64 · level recorded 2026-09-26
load_json in 344 papers preprocess.py ran MIT (permissive) · 7eb40d424f4306a5 · report fixture: signature data_dir:str/0/str, data_type:str/0/str · returned dict · level recorded 2026-09-26
option_postprocess utils/processing.py ran MIT (permissive) · 003e5dada7af2bae · report fixture: signature generated_inputs:list/1/str, data_type:str/0/str, is_logprob:bool/0/bool · returned tuple · level recorded 2026-09-26
preprocess_data in 44 papers utils/visualize.py ran MIT (permissive) · c501787fde562c3c · report fixture: signature json_data:dict/0/?, data_type:str/0/str, ordered_relations:list/1/str, new_ordered_relations:list/1/str · returned dict · level recorded 2026-09-26
prompt_fs_generator utils/prompts.py ran MIT (permissive) · 425c2228b8849976 · report fixture: signature dict:dict/0/?, model_type:str/0/str, data_type:str/0/str, template_type:int/0/int, shuffle:bool/0/bool, few_shot_k:int/0/int, few_shot_data_dir:str/0/str · returned str · level recorded 2026-09-26
prompt_generator utils/prompts.py ran MIT (permissive) · 6e27415d7d2bdc18 · report fixture: signature dict:dict/0/?, model_type:str/0/str, data_type:str/0/str, template_type:int/0/int, shuffle:bool/0/bool, prompt_variation:bool/0/bool · returned str · level recorded 2026-09-26
random_sample in 18 papers sample.py ran MIT (permissive) · 2163e5e3c4bcebcd · report fixture: signature data:dict/0/?, total_indices:list/1/str, sample_size:int/0/int, few_shot_k:int/0/int, seed:int/0/int · returned tuple · level recorded 2026-09-26
relative_entropy utils/metrics.py ran fingerprinted MIT (permissive) · 7e736694212e7c19 · report fixture: signature p:ndarray/1/float, q:ndarray/1/float · returned ndarray · level recorded 2026-09-26
str2bool in 533 papers preprocess.py ran · violated contract MIT (permissive) · 7c508037b40522af · report fixture: signature v:str/0/str · returned bool · level recorded, date not recorded
generate_openai in 6 papers utils/generation.py unverified MIT (permissive) · e5b0490b427152f2 · report
load_data in 700 papers preprocess.py unverified MIT (permissive) · 3bf33362230ed1a5 · report
load_model in 479 papers utils/models.py unverified MIT (permissive) · df8175311a751726 · report
preprocess_data_Nie utils/visualize.py unverified MIT (permissive) · 444acbf63a2aed61 · report
google-research/FLAN found in paper text by Syntology 10 samples, 10 ran (0 instrument-failed) · 0 unverified
collapse_related_tasks flan/task_splits.py ran Apache-2.0 (permissive) · 206789c8dcfa3c19 · report fixture: signature tasks:list/1/str, collapse_map:dict/0/? · returned list · level recorded 2026-09-26
enumerate_items flan/tasks.py ran Apache-2.0 (permissive) · fc3cbdfde1b3cc50 · report fixture: signature items_list:SymbolicTensor/1/<dtype: 'string'> · returned SymbolicTensor · level recorded 2026-09-26
format_from_feature_dictionary flan/preprocessors.py ran Apache-2.0 (permissive) · 48f4ea4298ef92ff · report fixture: signature format_string:str/0/str, feature_dictionary:dict/0/? · returned SymbolicTensor · level recorded 2026-09-26
is_superglue_task flan/task_splits.py ran Apache-2.0 (permissive) · f92e9e42a889c8dd · report fixture: signature task_name:str/0/str · returned bool · level recorded 2026-09-26
negate flan/preprocessors.py ran Apache-2.0 (permissive) · e435cf490873e109 · report fixture: signature dataset:DatasetV1Adapter/0/?, keys:list/1/str · returned DatasetV1Adapter · level recorded 2026-09-26
parse_glm_qa_answer flan/postprocessors.py ran Apache-2.0 (permissive) · 486ad3a49d33036e · report fixture: signature answer:str/0/str, example:dict/0/?, is_target:bool/0/bool · returned str · level recorded 2026-09-26
remove_input_patterns_options flan/utils.py ran fingerprinted Apache-2.0 (permissive) · 252eaeffae9c5998 · report fixture: signature input_pattern:str/0/str · returned str · level recorded 2026-09-26
remove_leading_quotes_and_spaces flan/postprocessors.py ran fingerprinted Apache-2.0 (permissive) · 7d7424269c38356a · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-26
t_name_to_flan_pattern_name flan/utils.py ran fingerprinted Apache-2.0 (permissive) · 0688ceac63ea3074 · report fixture: signature t_name:str/0/str · returned str · level recorded 2026-09-26
tokenize in 176 papers flan/preprocessors.py ran Apache-2.0 (permissive) · a0c365753489d524 · report fixture: signature dataset:DatasetV1Adapter/0/?, output_features:dict/0/?, keys:tuple/1/str, vocab_key:str/0/str · returned DatasetV1Adapter · level recorded 2026-09-26

Tasks

Natural Language InferenceNatural Language Understanding

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ALIGNfail

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections