Papers › Factual Confidence of LLMs: on Reliability and Robustness of Current Estimators

Factual Confidence of LLMs: on Reliability and Robustness of Current Estimators

19 Jun 2024arXiv:2406.13415archive 2025-07-28

Matéo Mahaut, Laura Aina, Paula Czarnowska, Momchil Hardalov, Thomas Müller, Lluís Màrquez

Large Language Models (LLMs) tend to be unreliable in the factuality of their answers. To address this problem, NLP researchers have proposed a range of techniques to estimate LLM's confidence over facts. However, due to the lack of a systematic comparison, it is not clear how the different methods compare to one another. To fill this gap, we present a survey and empirical comparison of estimators of factual confidence. We define an experimental framework allowing for fair comparison, covering both fact-verification and question answering. Our experiments across a series of LLMs indicate that trained hidden-state probes provide the most reliable confidence estimates, albeit at the expense of requiring access to weights and training data. We also conduct a deeper assessment of factual confidence by measuring the consistency of model behavior under meaning-preserving variations in the input. We find that the confidence of LLMs is often unstable across semantically equivalent inputs, suggesting that there is much room for improvement of the stability of models' parametric knowledge. Our code is available at (https://github.com/amazon-science/factual-confidence-of-llms).

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 16 code samples harvested from 1 repository linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 3 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 9 ran (6 with no instrument failure; 3 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

amazon-science/factual-confidence-of-llms officialmentioned on GitHubpytorchApache-2.0Syntology: 16 samples, 9 ran (3 instrument-failed) · 7 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

16 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 3 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
6ran
7unverified

Licence: 0 of the 16 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 amazon-science/factual-confidence-of-llms. “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.

amazon-science/factual-confidence-of-llms official repository 16 samples, 9 ran (3 instrument-failed) · 7 unverified
check_relation src/self_knowledge/data_gen/sf_graph_cleanup.py ran Apache-2.0 (permissive) · 64cac13c7712a758 · report fixture: signature graph:DataFrame/0/?, property_id:int/0/int, obj_id:str/0/str, subj_id:int/0/int · returned bool · level recorded 2026-09-24
compute_on_device src/self_knowledge/evaluation/consistency_utils.py ran · our draft was wrong Apache-2.0 (permissive) · 38bcad1abd42a39a · report fixture: signature input:tuple/1/int · returned tuple · level recorded 2026-09-24
extract_ids src/self_knowledge/data_gen/sf_graph_cleanup.py ran Apache-2.0 (permissive) · 96ff3da37ed8fb76 · report fixture: signature text:str/0/str, df_dataset:DataFrame/0/? · returned tuple · level recorded 2026-09-24
format_context_trex src/self_knowledge/data_gen/sentence_extractors.py ran · our draft was wrong Apache-2.0 (permissive) · f6be3e2860c3bb80 · report fixture: signature example:dict/0/? · returned dict · level recorded 2026-09-24
get_lpaqa_paraphrase src/self_knowledge/data_gen/sentence_extractors.py ran Apache-2.0 (permissive) · 5649462c5d6fa11b · report fixture: signature example:dict/0/?, lpaqa_folder:str/0/str, paraphrase_type:str/0/str · returned str · level recorded 2026-09-24
get_model in 368 papers src/self_knowledge/arch.py ran Apache-2.0 (permissive) · c6aa1425713fcf1e · report fixture: signature model_name:str/0/str, no_grad:bool/0/bool · returned tuple · level recorded 2026-09-24
get_tokenizer in 156 papers src/self_knowledge/arch.py ran · our draft was wrong Apache-2.0 (permissive) · 5bcc936c4c8aef88 · report fixture: signature model_name:str/0/str · returned Mock · level recorded 2026-09-24
sequence_log_score src/self_knowledge/evaluation/logit_scorer_utils.py ran Apache-2.0 (permissive) · a4baff91a17627e1 · report fixture: signature o:MockOutput/0/?, input_toks:Tensor/2/int, pad_token_id:int/0/int · returned Tensor · level recorded 2026-09-24
surrogate_logit_score src/self_knowledge/evaluation/logit_scorer_utils.py ran Apache-2.0 (permissive) · 4a8e5481004d91f3 · report fixture: signature o:MockGeneratorOutput/0/?, targets:Tensor/2/int, pad_token_id:int/0/int · returned Tensor · level recorded 2026-09-24
compute_all_scores src/self_knowledge/evaluation/consistency_utils.py unverified Apache-2.0 (permissive) · 94dd1ade2ee9847a · report
get_lama src/self_knowledge/data_gen/sampling.py unverified Apache-2.0 (permissive) · 5161229b89b4fad7 · report
get_nli_scoring_fun src/self_knowledge/evaluation/consistency_utils.py unverified Apache-2.0 (permissive) · 0e0da0e698e5347a · report
get_popQA src/self_knowledge/data_gen/sampling.py unverified Apache-2.0 (permissive) · 298e3e6ffc4b1483 · report
log_to_pandas_dataframe src/self_knowledge/slot_filling.py unverified Apache-2.0 (permissive) · 67ebd03e36b5d144 · report
lpaqa_paraphrases src/self_knowledge/data_gen/sentence_extractors.py unverified Apache-2.0 (permissive) · 33ab7dabeec5a3a2 · report
o_pop_matching src/self_knowledge/data_gen/sf_graph_cleanup.py unverified Apache-2.0 (permissive) · cfdd2155ad385a81 · report

Tasks

Fact VerificationQuestion Answering

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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