CoChat

GLP-1R–positive neurons in the lateral septum mediate the anorectic and weight-lowering effects of liraglutide in mice

Open access

The DOI matches its Crossref record; 58 of the 58 references it cites were found in a registry or an index.

Abstract

Liraglutide, a glucagon-like peptide-1 (GLP-1) analog, is approved for obesity treatment, but the specific neuronal sites that contribute to its therapeutic effects remain elusive. Here, we show that GLP-1 receptor-positive (GLP-1R-positive) neurons in the lateral septum (LSGLP-1R) play a critical role in mediating the anorectic and weight-loss effects of liraglutide. LSGLP-1R neurons were robustly activated by liraglutide, and chemogenetic activation of these neurons dramatically suppressed feeding. Targeted knockdown of GLP-1 receptors within the LS, but not in the hypothalamus, substantially attenuated liraglutide's ability to inhibit feeding and lower body weight. The activity of LSGLP-1R neurons rapidly decreased during naturalistic feeding episodes, while synaptic inactivation of LSGLP-1R neurons diminished the anorexic effects triggered by liraglutide. Together, these findings offer critical insights into the functional role of LSGLP-1R neurons in the physiological regulation of energy homeostasis and delineate their instrumental role in mediating the pharmacological efficacy of liraglutide.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1172/jci178239

Cites

58 of 58 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 GLP-1R–expressing neurons in the LS are activated by liraglutide. (A) Representative image showing c-Fos expression in the dorsal lateral septum (dLS) following i.p. injection of either saline or liraglutide. Scale bars: left, 200 μm; right, 50 μm. (B) Left: Quantification of GLP-1R+ neuron densities across various brain regions. Right: Quantification of c-Fos+ neurons in corresponding brain regions after i.p. injection of saline or liraglutide. VP, ventral pallidum; LH, lateral hypothalamic area; PVN, paraventricular hypothalamic nucleus; Arc, arcuate hypothalamic nucleus; DMH, dorsomedial hypothalamic nucleus; APN, anterior pretectal nucleus; Pn, pontine nuclei; DG, dentate gyrus; ECu, external cuneate nucleus; LRt, lateral reticular nucleus; AP, area postrema. (C) Representative image illustrating position of infusion cannula above GLP-1R–positive neurons. Scale bar: 500 μm. (D) Representative image showing c-Fos expression in LSGLP-1R neurons. Scale bar: 50 μm. (E) Quantification of c-Fos+ GLP-1R+ (left) and c-Fos+ GLP-1R– cells (right) after administration of saline or liraglutide. Unpaired, 2-tailed t test; left: t(5) = 3.454, P = 0.0182; right: t(5) = 0.2629, P = 0.8031. Means ± SEM. (F) Schematic of whole-cell recording from tdTomato–positive neurons in GLP-1R-ires-Cre:Ai14 slices. (G) Representative trace of spontaneous excitatory postsynaptic currents in LSGLP-1R neurons before and after liraglutide application. (H) Left: Cumulative amplitude probability plots. Inset and right: Pooled data; n = 7 cells from 2 animals; paired 2-tailed t test; t(6) = 4.017/1.256, P = 0.0070/0.2557. *P < 0.05, **P < 0.01.
Figure 1 GLP-1R–expressing neurons in the LS are activated by liraglutide. (A) Representative image showing c-Fos expression in the dorsal lateral septum (dLS) following i.p. injection of either saline or liraglutide. Scale bars: left, 200 μm; right, 50 μm. (B) Left: Quantification of GLP-1R+ neuron densities across various brain regions. Right: Quantification of c-Fos+ neurons in corresponding brain regions after i.p. injection of saline or liraglutide. VP, ventral pallidum; LH, lateral hypothalamic area; PVN, paraventricular hypothalamic nucleus; Arc, arcuate hypothalamic nucleus; DMH, dorsomedial hypothalamic nucleus; APN, anterior pretectal nucleus; Pn, pontine nuclei; DG, dentate gyrus; ECu, external cuneate nucleus; LRt, lateral reticular nucleus; AP, area postrema. (C) Representative image illustrating position of infusion cannula above GLP-1R–positive neurons. Scale bar: 500 μm. (D) Representative image showing c-Fos expression in LSGLP-1R neurons. Scale bar: 50 μm. (E) Quantification of c-Fos+ GLP-1R+ (left) and c-Fos+ GLP-1R– cells (right) after administration of saline or liraglutide. Unpaired, 2-tailed t test; left: t(5) = 3.454, P = 0.0182; right: t(5) = 0.2629, P = 0.8031. Means ± SEM. (F) Schematic of whole-cell recording from tdTomato–positive neurons in GLP-1R-ires-Cre:Ai14 slices. (G) Representative trace of spontaneous excitatory postsynaptic currents in LSGLP-1R neurons before and after liraglutide application. (H) Left: Cumulative amplitude probability plots. Inset and right: Pooled data; n = 7 cells from 2 animals; paired 2-tailed t test; t(6) = 4.017/1.256, P = 0.0070/0.2557. *P < 0.05, **P < 0.01.CC BY · PubMed Central
Figure 2 Knockdown of GLP-1Rs in the LS, rather than in the PVN, attenuates the food intake– and body weight–reducing effects of liraglutide. (A) Left: Schematic of the viral construct for CRISPR/Cas9–mediated GLP-1R knockdown. Right: Representative image demonstrating virus expression in LSGLP-1R neurons. iLS, intermediate lateral septum; vLS, ventral lateral septum. Scale bar: 500 μm. (B) Left: Fluorescence images of dLS depicting GLP-1R immunohistochemistry. Scale bar: 200 μm. Right: Quantification of GLP-1R+ neurons in control and GLP-1R knockdown mice (gray: control, n = 5; red: GLP-1R knockdown, n = 5). Unpaired, 2-tailed t test; t(8) = 8.952, P < 0.0001. Means ± SEM. (C–H) Attenuation of liraglutide’s anorectic and body weight–reducing effects following GLP-1R knockdown in the dLS on standard chow and a high-sucrose food. Statistical results are provided for varying dosages and durations. Food intake (2 hours, left panels; 24 hours, middle panels): 2-way repeated-measures ANOVA, followed by Šidák’s post hoc test. Body weight (right panels): unpaired, 2-tailed t test. Means ± SEM. (I) Scheme depicting the investigation of the long-term effect of GLP-1R knockdown in the dLS on liraglutide-induced weight loss. (J and K) Attenuation of the body weight–reducing effects of chronic systemic liraglutide during a standard chow (J) or high-fat food (HFF) diet (K) following GLP-1R knockdown in the dLS. Two-way repeated-measures ANOVA; chow: interaction: F(3,39) = 5.657, P = 0.0026; virus: F(1,13) = 10.90, P = 0.0057; HFF: interaction: F(3,48) = 7.646, P = 0.0003; virus: F(1,16) = 10.84, P = 0.0046. Šidák’s post hoc analysis, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Means ± SEM.
Figure 2 Knockdown of GLP-1Rs in the LS, rather than in the PVN, attenuates the food intake– and body weight–reducing effects of liraglutide. (A) Left: Schematic of the viral construct for CRISPR/Cas9–mediated GLP-1R knockdown. Right: Representative image demonstrating virus expression in LSGLP-1R neurons. iLS, intermediate lateral septum; vLS, ventral lateral septum. Scale bar: 500 μm. (B) Left: Fluorescence images of dLS depicting GLP-1R immunohistochemistry. Scale bar: 200 μm. Right: Quantification of GLP-1R+ neurons in control and GLP-1R knockdown mice (gray: control, n = 5; red: GLP-1R knockdown, n = 5). Unpaired, 2-tailed t test; t(8) = 8.952, P < 0.0001. Means ± SEM. (C–H) Attenuation of liraglutide’s anorectic and body weight–reducing effects following GLP-1R knockdown in the dLS on standard chow and a high-sucrose food. Statistical results are provided for varying dosages and durations. Food intake (2 hours, left panels; 24 hours, middle panels): 2-way repeated-measures ANOVA, followed by Šidák’s post hoc test. Body weight (right panels): unpaired, 2-tailed t test. Means ± SEM. (I) Scheme depicting the investigation of the long-term effect of GLP-1R knockdown in the dLS on liraglutide-induced weight loss. (J and K) Attenuation of the body weight–reducing effects of chronic systemic liraglutide during a standard chow (J) or high-fat food (HFF) diet (K) following GLP-1R knockdown in the dLS. Two-way repeated-measures ANOVA; chow: interaction: F(3,39) = 5.657, P = 0.0026; virus: F(1,13) = 10.90, P = 0.0057; HFF: interaction: F(3,48) = 7.646, P = 0.0003; virus: F(1,16) = 10.84, P = 0.0046. Šidák’s post hoc analysis, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Means ± SEM.CC BY · PubMed Central
Figure 3 Silencing of LSGLP-1R neurons enhances consumption and reduces liraglutide’s effects on food intake and body weight. (A) Representative image showing specific expression of TeNT-2A-EGFP in LSGLP-1R neurons. Scale bars: left, 200 μm; right, 50 μm. (B) Left: Images of EYFP- and TeNT-expressing mice after 6 weeks on high-fat diet. Right: Quantification of body weight gain in mice after high-fat diet. Two-way repeated-measures ANOVA; standard chow: F(1,15) = 7.203, P = 0.0170. Šidák’s post hoc analysis. Means ± SEM. (C and D) Left: Scheme depicting free consumption of a standard chow and a high-sucrose food paradigm. Right: Quantification of daily solid food intake in EYFP- and TeNT-expressing mice. Unpaired, 2-tailed t test; chow: t(29) = 0.1347, P = 0.8970; high-sucrose food: t(17) = 3.529, P = 3.529. Means ± SEM. (E) Oral glucose tolerance test (oGTT) results for TeNT- and EYFP-expressing mice. Two-way repeated-measures ANOVA; F(5,60) = 0.7675, P = 0.5769. Inset: Area under the curve (AUC) over 2 hours during oGTT. Unpaired, 2-tailed t test; t(12) = 1.352, P = 0.2014. Means ± SEM. (F) Schematic showing Ensure solution self-administration (fixed ratio 1) paradigm. (G) Lick patterns for individual mice. (H) The TeNT group had a greater number of pokes at the active port and greater Ensure solution consumption than the EYFP group. Two-way repeated-measures ANOVA; F(1,21) = 119.2, P < 0.0001. Šidák’s post hoc analysis. Unpaired, 2-tailed t test; t(21) = 13.32, P < 0.0001. Means ± SEM. (I–L) Silencing of LSGLP-1R neurons attenuated the anorectic and body weight–reducing effect of acutely delivered systemic liraglutide during standard chow or high-sucrose diet over varying durations and dosages. Food intake (2 hours, left panels; 24 hours, middle panels): 2-way repeated-measures ANOVA, followed by Šidák’s post hoc test. Body weight (right panels): unpaired, 2-tailed t test. Means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.
Figure 3 Silencing of LSGLP-1R neurons enhances consumption and reduces liraglutide’s effects on food intake and body weight. (A) Representative image showing specific expression of TeNT-2A-EGFP in LSGLP-1R neurons. Scale bars: left, 200 μm; right, 50 μm. (B) Left: Images of EYFP- and TeNT-expressing mice after 6 weeks on high-fat diet. Right: Quantification of body weight gain in mice after high-fat diet. Two-way repeated-measures ANOVA; standard chow: F(1,15) = 7.203, P = 0.0170. Šidák’s post hoc analysis. Means ± SEM. (C and D) Left: Scheme depicting free consumption of a standard chow and a high-sucrose food paradigm. Right: Quantification of daily solid food intake in EYFP- and TeNT-expressing mice. Unpaired, 2-tailed t test; chow: t(29) = 0.1347, P = 0.8970; high-sucrose food: t(17) = 3.529, P = 3.529. Means ± SEM. (E) Oral glucose tolerance test (oGTT) results for TeNT- and EYFP-expressing mice. Two-way repeated-measures ANOVA; F(5,60) = 0.7675, P = 0.5769. Inset: Area under the curve (AUC) over 2 hours during oGTT. Unpaired, 2-tailed t test; t(12) = 1.352, P = 0.2014. Means ± SEM. (F) Schematic showing Ensure solution self-administration (fixed ratio 1) paradigm. (G) Lick patterns for individual mice. (H) The TeNT group had a greater number of pokes at the active port and greater Ensure solution consumption than the EYFP group. Two-way repeated-measures ANOVA; F(1,21) = 119.2, P < 0.0001. Šidák’s post hoc analysis. Unpaired, 2-tailed t test; t(21) = 13.32, P < 0.0001. Means ± SEM. (I–L) Silencing of LSGLP-1R neurons attenuated the anorectic and body weight–reducing effect of acutely delivered systemic liraglutide during standard chow or high-sucrose diet over varying durations and dosages. Food intake (2 hours, left panels; 24 hours, middle panels): 2-way repeated-measures ANOVA, followed by Šidák’s post hoc test. Body weight (right panels): unpaired, 2-tailed t test. Means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.CC BY · PubMed Central
Figure 4 Food consumption rapidly inhibits the activity of dLSGLP-1R neurons, but food seeking has no effect. (A) Representative image displaying optic fiber position above GCaMP6-expressing LSGLP-1R neurons. Scale bar: 200 μm. (B) Raw trace illustrating real-time Ca2+ dynamics within dLSGLP-1R neurons of freely ambulating (moving) mice engaged in feeding. (C) Schematic diagram representing the food deprivation protocol used. (D and E) Comprehensive heatmaps (top) and per-animal stacked plots (bottom) depicting average GCaMP6 responses within dLSGLP-1R neurons, synchronized to the first and last bites during bouts of consumption across various food types (standard chow, high-sucrose, or high-fat) in individually fasted subjects (n = 10 animals). (F) Calculation of the valley and area under the curve (AUC) for GCaMP6 signals in individual subjects, assessed within a 10-second window immediately after the first bite. (G) Schematic illustration of the satiety protocol. (H and I) Heatmaps (top) and mean traces of GCaMP6 signals (bottom) from dLSGLP-1R neurons, correlating with the commencement (H) and cessation (I) of high-fat food ingestion in both fasted and sated mice (n = 7 animals). (J) Quantitative examination of the valley and AUC z scores associated with consumption-induced Ca2+ responses within dLSGLP-1R neurons across fasted and sated mice. (K) Stacked trace of dLSGLP-1R-GCaMP6 signals, aligned to the first bite during the initial third and the concluding third of bouts in mice, analyzed under both fasted and fed states (n = 7 animals). (L) Comparative analysis of the valley of dLSGLP-1R-GCaMP6 signals between initial and final feeding stages. (M) Stacked plot showcasing the response of dLSGLP-1R-GCaMP6 when fasted mice engaged in exploratory actions (n = 7 animals). *P < 0.05, **P < 0.01.
Figure 4 Food consumption rapidly inhibits the activity of dLSGLP-1R neurons, but food seeking has no effect. (A) Representative image displaying optic fiber position above GCaMP6-expressing LSGLP-1R neurons. Scale bar: 200 μm. (B) Raw trace illustrating real-time Ca2+ dynamics within dLSGLP-1R neurons of freely ambulating (moving) mice engaged in feeding. (C) Schematic diagram representing the food deprivation protocol used. (D and E) Comprehensive heatmaps (top) and per-animal stacked plots (bottom) depicting average GCaMP6 responses within dLSGLP-1R neurons, synchronized to the first and last bites during bouts of consumption across various food types (standard chow, high-sucrose, or high-fat) in individually fasted subjects (n = 10 animals). (F) Calculation of the valley and area under the curve (AUC) for GCaMP6 signals in individual subjects, assessed within a 10-second window immediately after the first bite. (G) Schematic illustration of the satiety protocol. (H and I) Heatmaps (top) and mean traces of GCaMP6 signals (bottom) from dLSGLP-1R neurons, correlating with the commencement (H) and cessation (I) of high-fat food ingestion in both fasted and sated mice (n = 7 animals). (J) Quantitative examination of the valley and AUC z scores associated with consumption-induced Ca2+ responses within dLSGLP-1R neurons across fasted and sated mice. (K) Stacked trace of dLSGLP-1R-GCaMP6 signals, aligned to the first bite during the initial third and the concluding third of bouts in mice, analyzed under both fasted and fed states (n = 7 animals). (L) Comparative analysis of the valley of dLSGLP-1R-GCaMP6 signals between initial and final feeding stages. (M) Stacked plot showcasing the response of dLSGLP-1R-GCaMP6 when fasted mice engaged in exploratory actions (n = 7 animals). *P < 0.05, **P < 0.01.CC BY · PubMed Central
Figure 5 Activation of LSGLP-1R neurons suppresses food consumption and lowers feeding motivation. (A) Representative illustration demonstrating specific hM3D-mCherry expression within LSGLP-1R neurons. Scale bars: left, 200 μm; right, 50 μm. (B) Left: Representative depiction of robust c-Fos expression. Right: Statistical comparison. Unpaired, 2-tailed t test; t(6)= 23.53, P < 0.0001. Means ± SEM. Scale bar: 50 μm. (C) Paradigm for analyzing food consumption. (D) Effects of CNO injection on reduction of standard chow or high-sucrose food intake specifically in hM3D-expressing mice but not in mCherry-expressing mice. Two-way repeated-measures ANOVA (standard chow, F(1,17) = 22.07, P = 0.0002; high-sucrose food, F(1,17) = 8.784, P = 0.0087) followed by Šidák’s post hoc test. Means ± SEM. (E) Oral glucose tolerance test (oGTT) results for hM3D- and mCherry-expressing mice. Two-way repeated-measures ANOVA; F(5,75) = 0.2521, P = 0.9375. Inset: AUC over 2 hours during oGTT. Unpaired, 2-tailed t test; t(15) = 0.06426, P = 0.9496. Means ± SEM. (F) Paradigm for analyzing long-term chronic effects of chemogenetic activation of LSGLP-1R neurons on food intake and body weight. (G) Prolonged activation of LSGLP-1R neurons suppresses food intake. Two-way repeated-measures ANOVA (F(1,10) = 14.19, P = 0.0037) followed by Šidák’s post hoc test. Means ± SEM. (H) Prolonged activation of LSGLP-1R neurons suppresses body weight. Two-way repeated-measures ANOVA (F(1,10) = 8.782, P = 0.0142) followed by Šidák’s post hoc test. Means ± SEM. (I) Representative image displaying specific ChR2-mCherry expression within LSGLP-1R neurons, accompanied by fiber tracks situated above. Scale bars: left, 500 μm; right, 50 μm. (J) Schematic diagram depicting the poke-based Ensure solution intake paradigm. (K) Optogenetic stimulation of LSGLP-1R neurons resulted in a decline in the number of pokes at the active port and in consumption of Ensure solution. Left: Two-way repeated-measures ANOVA; F(1,20) = 12.27, P = 0.0022. Šidák’s post hoc analysis. Right: Paired, 2-tailed t test; t(5) = 4.932, P = 0.0044. **P < 0.01, ***P < 0.001, ****P < 0.0001.
Figure 5 Activation of LSGLP-1R neurons suppresses food consumption and lowers feeding motivation. (A) Representative illustration demonstrating specific hM3D-mCherry expression within LSGLP-1R neurons. Scale bars: left, 200 μm; right, 50 μm. (B) Left: Representative depiction of robust c-Fos expression. Right: Statistical comparison. Unpaired, 2-tailed t test; t(6)= 23.53, P < 0.0001. Means ± SEM. Scale bar: 50 μm. (C) Paradigm for analyzing food consumption. (D) Effects of CNO injection on reduction of standard chow or high-sucrose food intake specifically in hM3D-expressing mice but not in mCherry-expressing mice. Two-way repeated-measures ANOVA (standard chow, F(1,17) = 22.07, P = 0.0002; high-sucrose food, F(1,17) = 8.784, P = 0.0087) followed by Šidák’s post hoc test. Means ± SEM. (E) Oral glucose tolerance test (oGTT) results for hM3D- and mCherry-expressing mice. Two-way repeated-measures ANOVA; F(5,75) = 0.2521, P = 0.9375. Inset: AUC over 2 hours during oGTT. Unpaired, 2-tailed t test; t(15) = 0.06426, P = 0.9496. Means ± SEM. (F) Paradigm for analyzing long-term chronic effects of chemogenetic activation of LSGLP-1R neurons on food intake and body weight. (G) Prolonged activation of LSGLP-1R neurons suppresses food intake. Two-way repeated-measures ANOVA (F(1,10) = 14.19, P = 0.0037) followed by Šidák’s post hoc test. Means ± SEM. (H) Prolonged activation of LSGLP-1R neurons suppresses body weight. Two-way repeated-measures ANOVA (F(1,10) = 8.782, P = 0.0142) followed by Šidák’s post hoc test. Means ± SEM. (I) Representative image displaying specific ChR2-mCherry expression within LSGLP-1R neurons, accompanied by fiber tracks situated above. Scale bars: left, 500 μm; right, 50 μm. (J) Schematic diagram depicting the poke-based Ensure solution intake paradigm. (K) Optogenetic stimulation of LSGLP-1R neurons resulted in a decline in the number of pokes at the active port and in consumption of Ensure solution. Left: Two-way repeated-measures ANOVA; F(1,20) = 12.27, P = 0.0022. Šidák’s post hoc analysis. Right: Paired, 2-tailed t test; t(5) = 4.932, P = 0.0044. **P < 0.01, ***P < 0.001, ****P < 0.0001.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 58 references

Cited by

Metadata

Authors
Chen, Z., Deng, X., Shi, C., Jing, H., Tian, Y., Zhong, J., Chen, G., Xu, Y., Luo, Y., & Zhu, Y.
Affiliations
  • Chinese Academy of Sciences (CN)
  • Hunan Normal University (CN)
  • City University of Hong Kong (HK)
Corresponding author
Zhu, Yingjie
Venue
Journal of Clinical Investigation
Published
2024-09-03
Volume
134
Issue
17
ISSN
  • 0021-9738 (ISSN-L)
  • 1558-8238
OpenAlex
W4402124375
Cited by
48
References
58
Citations by year
2026: 13 · 2025: 32 · 2024: 3
FWCI
10.56
Type
journal article
Language
en
Open access
Locations
  • Journal of Clinical Investigation (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • CityU Scholars (submittedVersion, cc-by)
Funders
  • Major Project of the Science and Technology Innovation 2030 of China (2021ZD0202103)
  • National Natural Science Foundation of China (82171492,32200827,82271534)
  • Department of Science and Technology of Guangdong Province (2023B1515040009,2024A1515011612)
  • Science, Technology and Innovation Commission of Shenzhen (RCJC20200714114556103,ZDSYS20190902093601675,JCYJ20210324141201003)
Topics
  • Life Sciences › Neuroscience › Endocrine and Autonomic Systems › Regulation of Appetite and Obesity
  • Health Sciences › Medicine › Pharmacology (medical) › Pharmacology and Obesity Treatment
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Treatment and Management
Keywords
liraglutide, GLP-1 receptor, lateral septum, anorectic effect, weight loss, obesity treatment, feeding behavior, energy homeostasis, chemogenetic activation, synaptic inhibition, GLP-1 analogues
MeSH terms
Glucagon-Like Peptide-1 Receptor, Liraglutide, Glucagon-Like Peptide-1 Receptor Agonists, Animals, Appetite Depressants, Male, Neurons, Septal Nuclei, Weight Loss, Mice

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.

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

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.