Evaluating the efficacy of repetitive anodal transcranial direct current stimulation on cognitive fatigue in long COVID: A randomized controlled trial
Open accessAbstract
The enduring and debilitating symptoms of long COVID, associated with long-term recovery from SARS-CoV-2 infection have emerged as a significant public health concern in the aftermath of the global pandemic [1]. Among these symptoms, cognitive fatigue stands out due to its widespread occurrence and its significant effects on patients' quality of life [2]. In light of this emerging condition, the demand for effective treatment options has become imperative, prompting researchers and clinicians to seek solutions [3].
AI snapshot
Written from the paper’s full text.
TL;DR
In a pre-registered randomized, double-blind, sham-controlled trial of 40 long COVID patients with cognitive fatigue, four daily sessions of anodal tDCS over the left dlPFC were feasible and safe but did not produce stimulation-specific improvements in subjective cognitive fatigue. Both active and sham groups showed significant reductions in MFIS cognitive fatigue, depressive symptoms, and some quality-of-life measures up to one month, while EEG markers differed by condition.
Key findings
- Primary subjective cognitive fatigue improved across all participants, not specifically the active tDCS group: MFIS cognitive pre-to-post mean difference = 2.38, p < .001; pre-to-follow-up mean difference = 3.35, p < .001; post-to-follow-up mean difference = 0.98, p = .095.
- Depressive symptoms and quality of life improved across all participants: BDI-II pre-to-post mean difference = 1.6, p = .003 and pre-to-follow-up = 2.0, p < .001; EQ-5D-5L health scale pre-to-follow-up mean difference = −5.40, p = .046; health index pre-to-post = −0.09, p = .003 and pre-to-follow-up = −0.07, p = .010.
- EEG showed stimulation-specific modulation: occipital alpha power increased from post-assessment to follow-up only in the sham group (sham mean difference = −0.09, p = .003; verum = −0.02, p = .483); frontal theta showed a significant increase only in the sham group (sham = −0.04, p < .001; verum = −0.01, p = .283).
- Safety and feasibility: participants experienced minimal side effects during the four stimulation sessions, and no significant adverse events were reported.
Methods
- Pre-registered (DRKS00031294) randomized, double-blind, sham-controlled trial in 40 long COVID patients with cognitive fatigue.
- Intervention: repetitive anodal tDCS over the left dorsolateral prefrontal cortex; four consecutive daily sessions of 30 minutes active (1.5 mA) or sham stimulation, with assessments before treatment, one day post-treatment, and at one-month follow-up.
- Primary outcome: subjective cognitive trait fatigue measured by the Modified Fatigue Impact Scale (MFIS) cognitive subscore; additional assessments included depressive symptoms, quality of life, and EEG for objective trait fatigue. Statistical analysis details are in the Supplementary Material; the letter reports mean differences and p-values.
Limitations
- Stated: the study did not reveal stimulation-specific effects on subjective cognitive fatigue; improvements in both groups may reflect placebo effects driven by interpersonal dynamics and participant expectations that could overshadow tDCS-specific effects.
- Stated: future research should improve stimulation parameters, examine contextual factors, and evaluate potential interactions between tDCS and other treatment approaches.
- Evident from the text: the small sample (n = 40) and short four-session intervention limit power and generalizability; statistical analysis details are relegated to supplementary material, and the letter does not report effect sizes or between-group statistics.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
8 of 9 cited sources checked · 3 supported · 0 flagged · accuracy not yet measured
- Premisecitation§ Dear Editor,
Placebo effects driven by interpersonal dynamics and participant expectations could overshadow tDCS-specific effects.
“which highlight placebo effects driven by interpersonal dynamics and participant expectations, that could overshadow tDCS-specific effects [10].”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites [10] Linnhoff, 2023, Effects of repetitive twice-weekly transcranial direct current stimulations on fatigue and fatigability in people with multiple sclerosis, Sci Rep source
The Discussion’s “Placebo effect” section links the observed placebo effect to positive experimenter–participant interactions and reports that placebo effects may partially mask anodal tDCS effects. It does not identify participant expectations as a driver of those effects. (LLM, conf=0.85)
Experimental — accuracy not yet measured
- Findingown data§ Dear Editor,
Across all participants, there were significant reductions in depressive symptom ratings from pre- to post-assessment (mean difference = 1.6, p = .003).
“We found significant reductions in depressive symptom ratings from pre-to post-assessment (mean difference = 1.6, p = .003)”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
The improvements in MFIS cognitive scores were not exclusive to the active stimulation group.
“Although these improvements were not exclusive to the active stimulation group, it is essential to highlight that both groups experienced substantial reductions in their so far treatment-resistant cognitive fatigue.”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
We identified long-term effects on electrophysiological markers related to state fatigue after four daily sessions of tDCS.
“While general improvements were observed in subjective state fatigue, depression and quality of life, we identified long-term effects on electrophysiological markers related to state fatigue after four daily sessions of tDCS.”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
Across all participants, there was a non-significant reduction in subjective trait fatigue ratings from post-assessment to follow-up (mean difference = 0.98, p = .095).
“A further, non-significant reduction was noted from post-assessment to follow-up (mean difference = 0.98, p = .095).”
Quote in the paperNot a substantial claim
- Findingown data§ Dear Editor,
Improvements in subjective health as measured by EQ-5D-5L were observed in both groups.
“Improvements in depressive symptoms as measured by BDI-II scores and subjective health measured by EQ-5D-5L (Fig. 1Cii -1Civ) were observed in both groups.”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
The study did not reveal stimulation-specific effects on subjective cognitive fatigue.
“Although the study did not reveal stimulation-specific effects on subjective cognitive fatigue, the overall improvements observed in both groups emphasize the potential of tDCS as a component of a comprehensive treatment strategy for long-term post-COVID fatigue.”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
There was a significant increase in mean occipital alpha power from post-assessment to follow-up exclusively in the sham group (mean difference sham = -0.09, p = .003; verum = -0.02, p = .483).
“We observed a significant increase in mean occipital alpha power from postassessment to follow-up exclusively in the sham group (mean difference sham = -0.09,p = .003;verum = -0.02,p = .483).”
Quote in the paperNot a substantial claim - Findingown data§ Dear Editor,
Across all participants, there was a statistically significant reduction in subjective trait fatigue ratings from pre- to post-assessment (mean difference = 2.38, p < .001).
“We observed a statistical significant reduction in subjective trait fatigue ratings from pre-to post-assessment (mean difference = 2.38, p < .001)”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
EEG data revealed a stimulation-specific modulation in task-related occipital alpha and frontal theta activity.
“The EEG data revealed a stimulation specific modulation in taskrelated occipital alpha and frontal theta activity (Fig. 1D).”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
Improvements in depressive symptoms as measured by BDI-II scores were observed in both groups.
“Improvements in depressive symptoms as measured by BDI-II scores and subjective health measured by EQ-5D-5L (Fig. 1Cii -1Civ) were observed in both groups.”
Quote in the paperNot a substantial claim - Findingown data§ Dear Editor,
The improvements in MFIS cognitive scores lasted up to one month following the intervention.
“The results revealed significant improvements in MFIS cognitive scores for both, the active and sham stimulation group, with these benefits lasting up to one month following the intervention (Fig. 1Ci).”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
Both the active and sham stimulation groups showed significant improvements in MFIS cognitive scores.
“The results revealed significant improvements in MFIS cognitive scores for both, the active and sham stimulation group, with these benefits lasting up to one month following the intervention (Fig. 1Ci).”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
Both groups experienced substantial reductions in their so far treatment-resistant cognitive fatigue.
“Although these improvements were not exclusive to the active stimulation group, it is essential to highlight that both groups experienced substantial reductions in their so far treatment-resistant cognitive fatigue.”
Quote in the paperSubstantial claim- Rests on an assumption: The cognitive fatigue in the study participants was treatment-resistant prior to the intervention.
- Findingown data§ Dear Editor,
Across all participants, EQ-5D-5L health index ratings increased from pre- to post-assessment (mean difference = -0.09, p = .003).
“Also, health index ratings increased from pre-to post-assessment (mean difference = -0.09,p = .003)”
Quote in the paperSubstantial claim - Findingown data§ Dear Editor,
Across all participants, depressive symptom ratings remained stable from post-assessment to follow-up (mean difference = 0.4, p = .428).
“Ratings remained stable from postassessment to follow-up (mean difference = 0.4, p = .428).”
Quote in the paperNot a substantial claim - Findingown data§ Dear Editor,
Across all participants, there was a statistically significant reduction in subjective trait fatigue ratings from pre-assessment to follow-up (mean difference = 3.35, p < .001).
“and from pre-assessment to follow-up (mean difference = 3.35, p < .001)across all participants.”
Quote in the paperSubstantial claim - Interpretationown data§ Dear Editor,
The benefits of the intervention extend beyond mere fatigue reduction, potentially addressing the wider implications of long COVID on patients' overall well-being.
“These findings suggest that the benefits of the intervention extend beyond mere fatigue reduction, potentially addressing the wider implications of long COVID on patients' overall well-being.”
Quote in the paperSubstantial claim - Interpretationown data§ Dear Editor,
There are distinct effects on electrophysiological markers of fatigue between the active and sham conditions.
“These findings suggest distinct effects on electrophysiological markers of fatigue between the active and sham conditions.”
Quote in the paperSubstantial claim
- Classified as probably not a substantial claim. 6
- May combine two claims. 8
Classified as probably not a substantial claim.
Premisecitation§ Dear Editor,Non-invasive brain stimulation via tDCS has been proposed as a viable solution.
“Non-invasive brain stimulation via transcranial direct current stimulation (tDCS) has been proposed as a viable solution [4][5][6].”
Quote in the paperSubstantial claim3 of 3 sources checked- cites [4] Cavendish, 2022, Combination of transcranial direct current stimulation with online cognitive training improves symptoms of Post-acute Sequelae of COVID-19: a case series, Brain Stimul source
- cites [5] Pinto, 2023, Prefrontal tDCS modulates autonomic responses in COVID-19 inpatients, Brain Stimul source
- cites [6] Andrade, 2022, Efficacy and safety of HD-tDCS and respiratory rehabilitation for critically ill patients with COVID-19 the HD-RECOVERY randomized clinical trial, Brain Stimul source
May combine two claims.
Premisecitation§ Dear Editor,tDCS may counteract the frontal hypoactivity observed in long COVID patients and therefore impact the cortico-striato-thalamo-cortical network, a circuit closely linked to fatigue.
“tDCS may counteract the frontal hypoactivity observed in long COVID patients and therefore impact the cortico-striato-thalamo-cortical networka circuit closely linked to fatigue [7,8].”
Quote in the paperSubstantial claim2 of 2 sources checked- cites [7] Santana, 2023, Non-invasive brain stimulation for fatigue in post-acute sequelae of SARS-CoV-2 (PASC), Brain Stimul source
The Discussion (page not provided) says frontal and central areas are dysfunctional after COVID-19 and that HD-tDCS can affect cortical and subcortical circuits. The Introduction and Conclusion link motor-cortex stimulation to improvement in PASC-related fatigue. The paper does not specify frontal hypoactivity or identify the cortico-striato-thalamo-cortical network as the circuit affected. (LLM, conf=0.84)
Experimental — accuracy not yet measured - cites [8] Greene, 2024, Blood–brain barrier disruption and sustained systemic inflammation in individuals with long COVID-associated cognitive impairment, Nat Neurosci source
The Results section (“Long COVID-associated brain fog induces structural changes in the brain”; page unavailable) reports reduced frontal cortical thickness in the long-COVID and brain-fog cohorts, and the Discussion identifies fatigue as a long-COVID symptom. Those findings do not establish frontal hypoactivity, a link between the specified circuit and fatigue, or an effect of tDCS on that circuit. (LLM, conf=0.78)
Experimental — accuracy not yet measured
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Premisecitation§ Dear Editor,Cognitive fatigue stands out due to its widespread occurrence and its significant effects on patients' quality of life.
“Among these symptoms, cognitive fatigue stands out due to its widespread occurrence and its significant effects on patients' quality of life [2].”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites [2] Ceban, 2022, Fatigue and cognitive impairment in Post-COVID-19 Syndrome: a systematic review and meta-analysis, Brain Behav Immun source
The Discussion reports that persistent fatigue affects approximately a third of people after COVID-19 and cognitive impairment over a fifth, and associates these symptoms with marked functional impairment (page unavailable). The Results’ Functional Outcomes/Quality of Life section reports links to activity difficulties and lower quality-of-life scores, but the paper assesses fatigue and cognitive impairment separately; it does not establish the prevalence or quality-of-life effects of a distinct symptom called “cognitive fatigue.” (LLM, conf=0.84)
Experimental — accuracy not yet measured
Probably not a substantial claim: may not pass S3 (load-bearing).
Interpretationcitation§ Dear Editor,Our findings align with prior studies on neuromodulation.
“Yet, our findings align with prior studies on neuromodulation, which highlight placebo effects driven by interpersonal dynamics and participant expectations, that could overshadow tDCS-specific effects [10].”
Quote in the paperNot a substantial claim1 of 1 sources checked- cites [10] Linnhoff, 2023, Effects of repetitive twice-weekly transcranial direct current stimulations on fatigue and fatigability in people with multiple sclerosis, Sci Rep source
The Discussion’s “Placebo effect” section says that positive experimenter–participant interactions likely promoted a stimulation-independent placebo effect. It also reports that placebo effects in a prior tDCS study may have partially masked anodal effects. The paper does not identify participant expectations as a driver, so that part of the attributed claim is unsupported. (LLM, conf=0.85)
Experimental — accuracy not yet measured
May combine two claims.
Findingown data§ Dear Editor,There was a trend for frontal theta activity, with a significant increase from post-assessment to follow-up observed only in the sham group (mean difference sham = -0.04, p < .001; verum = -0.01, p = .283).
“Analogously, a trend was noted for frontal theta activity, with a significant increase from post-assessment to follow-up observed only in the sham group (mean difference sham = -0.04,p < .001;verum = -0.01,p = .283).”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown data§ Dear Editor,The study demonstrated the feasibility and safety of anodal tDCS as a treatment for fatigue associated with long COVID.
“Finally, it is crucial to highlight that our research has demonstrated the feasibility and safety of anodal tDCS as a treatment for fatigue associated with long COVID.”
- Also stated in: § Dear Editor, · plain
“In conclusion, this study further supports the safety and tolerability of repetitive tDCS among patients with long COVID.”
Quote in the paperSubstantial claim- Also stated in: § Dear Editor, · plain
May combine two claims.
Findingown data§ Dear Editor,During the four stimulation sessions, participants experienced minimal side effects, and no significant adverse events were reported.
“During the four stimulation sessions, participants experienced minimal side effects, and no significant adverse events were reported (Fig. 1B).”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
Study design and conductown data§ Dear Editor,The study was pre-registered (DRKS00031294), randomized, double-blind, and sham-controlled.
“In this pre-registered (DRKS00031294) randomized, double-blind and sham-controlled trial we assessed the efficacy of repetitive anodal tDCS over the left dorsolateral prefrontal cortex (dlPFC) in 40 long COVID patients with cognitive fatigue.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown data§ Dear Editor,The consistent improvements in cognitive trait fatigue, depressive symptoms, and quality of life highlight the clinical significance of these findings.
“The consistent improvements in cognitive trait fatigue, depressive symptoms, and quality of life highlight the clinical significance of these findings, particularly in light of the current scarcity of effective treatments for this patient demographic.”
Quote in the paperNot a substantial claim- Rests on an assumption: There is a current lack of effective treatments for long-term fatigue associated with long COVID.
May combine two claims.
Interpretationown data§ Dear Editor,This study demonstrates the feasibility, safety, and potential efficacy of repetitive anodal tDCS over the DLPFC in reducing cognitive fatigue.
“This study contributes to the advancement of this object by demonstrating the feasibility, safety, and potential efficacy of repetitive anodal tDCS over the DLPFC in reducing cognitive fatigue.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
Interpretationown data§ Dear Editor,The overall improvements observed in both groups emphasize the potential of tDCS as a component of a comprehensive treatment strategy for long-term post-COVID fatigue.
“Although the study did not reveal stimulation-specific effects on subjective cognitive fatigue, the overall improvements observed in both groups emphasize the potential of tDCS as a component of a comprehensive treatment strategy for long-term post-COVID fatigue.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Premisecitation§ Dear Editor,The enduring and debilitating symptoms of long COVID have emerged as a significant public health concern.
“The enduring and debilitating symptoms of long COVID, associated with long-term recovery from SARS-CoV-2 infection have emerged as a significant public health concern in the aftermath of the global pandemic [1].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [1] Al-Aly, 2024, Long COVID science, research and policy, Nat Med source
The paper relies on this without support.
The cognitive fatigue in the study participants was treatment-resistant prior to the intervention.
“both groups experienced substantial reductions in their so far treatment-resistant cognitive fatigue.”
The paper relies on this without support.
There is a current lack of effective treatments for long-term fatigue associated with long COVID.
“This finding is noteworthy given the current lack of effective treatments for long-term fatigue associated with long COVID.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Conclusion calls trait-fatigue outcomes state-fatigue outcomes
MinorThe methods identify the MFIS measure as subjective trait fatigue and the EEG measure as objective trait fatigue, but the conclusion describes the improvements and electrophysiological markers as related to state fatigue.
“While general improvements were observed in subjective state fatigue, depression and quality of life, we identified long-term effects on electrophysiological markers related to state fatigue after four daily sessions of tDCS.”
“The primary outcome measure was subjective cognitive trait fatigue, assessed through the Modified Fatigue Impact Scale (MFIS) cognitive subscore.Assessments additionally included the assessment of depressive symptoms, quality of life as well as an EEG measurement to capture objective trait fatigue [9].”
Cites
EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis
Klimesch, W. (1999) · Brain Research Reviews · doi:10.1016/s0165-0173(98)00056-3 · Cited by 6,889 · see citing works
Fatigue and cognitive impairment in Post-COVID-19 Syndrome: a systematic review and meta-analysis
Ceban, F., Ling, S., Lui, L. M. W., Lee, Y., Gill, H., Teopiz, K. M., Rodrigues, N. B., Subramaniapillai, M., Di Vincenzo, J. D., Cao, B., et al. (2022) · Brain, Behavior, and Immunity · doi:10.1016/j.bbi.2021.12.020 · Cited by 1,554 · see citing works
Long COVID science, research and policy
Al-Aly, Z., Davis, H., McCorkell, L., Soares, L., Wulf-Hanson, S., Iwasaki, A., & Topol, E. J. (2024) · Nature Medicine · doi:10.1038/s41591-024-03173-6 · Cited by 538 · see citing works
Greene, C., Connolly, R., Brennan, D., Laffan, A., O’Keeffe, E., Zaporojan, L., O’Callaghan, J., Thomson, B., Connolly, E., Argue, R., et al. (2024) · Nature Neuroscience · doi:10.1038/s41593-024-01576-9 · Cited by 429 · see citing works
Non-invasive brain stimulation for fatigue in post-acute sequelae of SARS-CoV-2 (PASC)
Santana, K., França, E., Sato, J., Silva, A., Queiroz, M., de Farias, J., Rodrigues, D., Souza, I., Ribeiro, V., Caparelli-Dáquer, E., et al. (2023) · Brain Stimulation · doi:10.1016/j.brs.2023.01.1672 · Cited by 52 · see citing works
Linnhoff, S., Koehler, L., Haghikia, A., & Zaehle, T. (2023) · Frontiers in Immunology · doi:10.3389/fimmu.2022.935614 · Cited by 32 · see citing works
Self-citationAndrade, S. M., Cecília de Araújo Silvestre, M., Tenório de França, E. É., Bezerra Sales Queiroz, M. H., de Jesus Santana, K., Lima Holmes Madruga, M. L., Torres Teixeira Mendes, C. K., Araújo de Oliveira, E., Bezerra, J. F., Barreto, R. G., et al. (2022) · Brain Stimulation · doi:10.1016/j.brs.2022.05.006 · Cited by 30 · see citing works
Cavendish, B. A., Lima, A., Bertola, L., Charvet, L., Bikson, M., Brunoni, A. R., & Vidal, K. S. (2022) · Brain Stimulation · doi:10.1016/j.brs.2022.09.008 · Cited by 18 · see citing works
Prefrontal tDCS modulates autonomic responses in COVID-19 inpatients
Pinto, T. P., Inácio, J. C., de Aguiar, E., Ferreira, A. S., Sudo, F. K., Tovar-Moll, F., & Rodrigues, E. C. (2023) · Brain Stimulation · doi:10.1016/j.brs.2023.03.001 · Cited by 11 · see citing works
Linnhoff, S., Haghikia, A., & Zaehle, T. (2023) · Scientific Reports · doi:10.1038/s41598-023-32779-y · Cited by 11 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Evaluation of Interventions for Cognitive Symptoms in Long COVID
Knopman, D. S., Attix, D. K., Laskowitz, D. T., Becker, J., Charvet, L. E., Wisnivesky, J. P., Federman, A. D., Silverstein, A. M., Lokhnygina, Y. V., Pilloni, G., et al. (2025) · JAMA Neurology · doi:10.1001/jamaneurol.2025.4415 · Cited by 11 · see citing works
Mischke, M., & Zaehle, T. (2026) · BMC Neuroscience · doi:10.1186/s12868-025-00989-x · Cited by 2 · see citing works
Self-citationRathinam, C., Heath, G., Gupta, R., Wassmer, E., McElroy, E., Horn, H., & Fernández‐Espejo, D. (2026) · Health Expectations · doi:10.1111/hex.70706
- Authors
- Mischke, M., & Zaehle, T.
- Affiliations
- Deutsches Zentrum für Psychische Gesundheit (DE)
- Center for Behavioral Brain Sciences (DE)
- Corresponding author
- Zaehle, Tino
- Venue
- Brain Stimulation
- Publisher
- Elsevier BV
- Published
- 2025-05
- Volume
- 18
- Issue
- 3
- Pages
- 733-735
- ISSN
- 1876-4754 (ISSN-L)
- 1935-861X
- OpenAlex
- W4408508904
- Semantic Scholar
- 2576b91a405b1abc611417eed0d980d24e7d3b2d
- Cited by
- 3
- References
- 10
- Citations by year
- 2026: 2 · 2025: 1
- FWCI
- 0.92
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Brain stimulation (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- State of Saxony Anhalt Ministry of Economics and Science
- Topics
- Life Sciences › Neuroscience › Neurology › Transcranial Magnetic Stimulation Studies
- Health Sciences › Medicine › Neurology › Long-Term Effects of COVID-19
- Health Sciences › Medicine › Epidemiology › Traumatic Brain Injury Research
- Keywords
- long COVID, mental fatigue, non-invasive brain stimulation, SARS-CoV-2, treatment efficacy, quality of life
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-28.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-28.
Cited sources are checked by gpt-6-sol.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
