The graphical brain: Belief propagation and active inference
Open accessThe DOI matches its Crossref record; 112 of the 112 references it cites were found in a registry or an index; none of the 73 sources its claims cite is checked for support yet.
Abstract
This paper considers functional integration in the brain from a computational perspective. We ask what sort of neuronal message passing is mandated by active inference—and what implications this has for context-sensitive connectivity at microscopic and macroscopic levels. In particular, we formulate neuronal processing as belief propagation under deep generative models. Crucially, these models can entertain both discrete and continuous states, leading to distinct schemes for belief updating that play out on the same (neuronal) architecture. Technically, we use Forney (normal) factor graphs to elucidate the requisite message passing in terms of its form and scheduling. To accommodate mixed generative models (of discrete and continuous states), one also has to consider link nodes or factors that enable discrete and continuous representations to talk to each other. When mapping the implicit computational architecture onto neuronal connectivity, several interesting features emerge. For example, Bayesian model averaging and comparison, which link discrete and continuous states, may be implemented in thalamocortical loops. These and other considerations speak to a computational connectome that is inherently state dependent and self-organizing in ways that yield to a principled (variational) account. We conclude with simulations of reading that illustrate the implicit neuronal message passing, with a special focus on how discrete (semantic) representations inform, and are informed by, continuous (visual) sampling of the sensorium.
Overview
CoChat Summary
TL;DR
This paper derives belief propagation under deep generative models as a computational architecture for functional integration in the brain, using Forney factor graphs to link discrete and continuous state representations. It maps the resulting message passing onto intrinsic/extrinsic cortical microcircuits and cortico–basal ganglia–thalamic loops, illustrating the scheme with simulations of pictographic reading.
Key findings
- Belief propagation under deep generative models can be expressed with Forney factor graphs; discrete models use A/B/C/D matrices and policies, continuous models use generalized coordinates and predictive coding, and mixed models require link factors implementing Bayesian model averaging and comparison.
- Discrete and continuous representations are linked by Bayesian model averaging (descending empirical priors) and Bayesian model comparison (ascending posterior over outcome models); thalamocortical loops may implement this, while expected free energy and policy expectations involve cortico–basal ganglia–thalamic loops.
- In simulations of reading, the model has two discrete hierarchical levels with six sentences and four words, with letters/pictograms in quadrants; saccades are assumed to take about 256 ms, continuous integration uses 24/25 time steps of about 10 ms, and the subject resolves the sentence on the last saccade of the last word.
- Minimizing variational free energy also minimizes Lyapunov exponents, so belief propagation organizes toward critical slowing; self-organized criticality is described as an epiphenomenon of self-evidencing.
Methods
- Theoretical/computational derivation using Bayesian networks, Forney factor graphs, and neural network/circuit mappings; derives belief propagation for discrete, continuous, and mixed generative models.
- Simulation of pictographic reading: two discrete hierarchical levels (six sentences, four words; letters/pictograms in quadrants), a continuous visuomotor model with 32 × 32 pixel images, equations integrated with 16 iterations per time point per discrete level, saccades about 256 ms, continuous integration 24/25 time steps of about 10 ms, and roughly 4.5 s of simulated time.
- Main analysis is simulation-based: eye movements, posterior probabilities/evidence accumulation, simulated firing rates/depolarization, and band-pass filtered local field potentials between 4 Hz and 32 Hz; no empirical dataset is analyzed.
Limitations
- Stated: the neural networks “should not be taken too seriously” and are offered as a starting point for refinement and deconstruction; many anatomical issues are ignored, including direct/indirect pathways and dopamine modulation.
- Stated: the sensory attenuation model is “quite a crude way” of modeling sensory attenuation, and the reading simulation is “a bit abstract.”
- Evident: the anatomical mappings are speculative and not empirically tested here, and the simulation results depend on the model assumptions and message-passing schedule.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures














Verification
Claims the paper makes
0 Valid · 1 Not valid · 45 Not checked · 146 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Here, the subject correctly infers, on the last saccade of the last word, that the first sentence generated the stimuli.”
§ The graphical brain: Belief propagation and active inference > SIMULATIONS OF READING > Categorization and Evidence Accumulation
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This means that a partially observed sequence of outcomes can inform expectations about the future, which are necessary to evaluate the expected free energy of a policy.”
§ The graphical brain: Belief propagation and active inference > DISCRETE GENERATIVE MODELS > Belief Updating and Propagation
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
None of the sources cited for this claim have been checked yet.
In the paper
“When a generative model is specified as a factor graph, latent variables can often be inferred by executing a message passing schedule that can be derived automatically.”
§ The graphical brain: Belief propagation and active inference > INTRODUCTION
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A statement the paper makes itself — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Factor graphs are an important type of probabilistic graphical model because they facilitate the derivation of (approximate) Bayesian inference algorithms.”
§ The graphical brain: Belief propagation and active inference > INTRODUCTION
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, note that the outcomes that inform the expected free energy are not the observed outcomes but predicted outcomes based upon expected states, under each policy (i.e., message 5).”
§ The graphical brain: Belief propagation and active inference > DISCRETE GENERATIVE MODELS
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
112 more claims
Create a free account to see all 192 claims and to verify the 73 references still open.
- Probably not a substantial claim. 39
- It may state two separate claims as one. 42
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“This means that systems able to make such future-oriented inferences must possess a generative model with temporal or counterfactual depth [26,39].”
From The Markov blankets of life: autonomy, active inference and the free energy principle (2018) · cites this paper as [39]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“In some simulations, one can mix discrete and continuous state-space models by placing the former on top of the latter, to produce deep generative models that, through active inference, can be used to simulate many known aspects of computational anatomy and physiology in the brain [158].”
From The free energy principle made simpler but not too simple (2023) · cites this paper as [158]
The cited source says what the claim attributes to it.
In the citing paper
“In this sense, it is sometimes referred to as epistemic affordance [158].”
From The free energy principle made simpler but not too simple (2023) · cites this paper as [158]
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.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ The graphical brain: Belief propagation and active inference > MIXED MODELS
“E(t)m=−lnoτ,m−∫0TL(t)mdtL(t)m=lnP(õ(t)|ηm)−lnP(õ(t)|η).(12)”
§ The graphical brain: Belief propagation and active inference > MIXED MODELS
“This equation expresses the free energy E of competing outcome models as the prior surprise plus the log evidence for each outcome model (integrated over time).”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ The graphical brain: Belief propagation and active inference > SIMULATIONS OF READING > Categorization and Evidence Accumulation
“This figure illustrates belief propagation (producing the behavior shown in Figure 11), in terms of simulated firing rates and depolarization.”
§ The graphical brain: Belief propagation and active inference > SIMULATIONS OF READING
“Figure 12 shows the behavior that results from integrating the message passing scheme described in Figure 10.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ The graphical brain: Belief propagation and active inference > SIMULATIONS OF READING
“each of the 25 time steps of the continuous integration scheme corresponds roughly to 10 ms”
§ The graphical brain: Belief propagation and active inference > SIMULATIONS OF READING
“This trajectory is shown as a function of the time steps used to integrate the generalized belief updating (24 time steps, each corresponding to about 10 ms of real time).”
Citations
Cites
(1988) · Cited by 16,582
Distributed hierarchical processing in the primate cerebral cortex
Felleman, D. J., & Van Essen, D. C. (1991) · Cerebral Cortex · doi:10.1093/cercor/1.1.1 · Cited by 7,783 · cited 2×
“Given that the message passing requires prediction errors on hidden causes to be passed to higher levels, one can assume they are encoded by superficial pyramidal cells; that is, the source of ascending extrinsic connections (Bastos et al., 2012; Felleman & Van Essen, 1991; Markov et al., 2013).”
BackgroundSelf-organized criticality: An explanation of the 1/f noise
Bak, P., Tang, C., & Wiesenfeld, K. (1987) · Physical Review Letters · doi:10.1103/physrevlett.59.381 · Cited by 7,591 · cited 1×
“From the perspective of active inference, this implies self-evidencing (Hohwy, 2016), which entails self-organized criticality (Bak et al., 1987) and dynamic connectivity (Allen et al., 2012; Breakspear, 2004).”
BackgroundFactor graphs and the sum-product algorithm
Kschischang, F. R., Frey, B. J., & Loeliger, H.-A. (2001) · IEEE Transactions on Information Theory · doi:10.1109/18.910572 · Cited by 6,486
Whatever next? Predictive brains, situated agents, and the future of cognitive science
Clark, A. (2013) · Behavioral and Brain Sciences · doi:10.1017/s0140525x12000477 · Cited by 5,924 · cited 1×
“In particular, we assume that the brain engages in some form of (Bayesian) inference—and can therefore be described as maximizing Bayesian model evidence (Clark, 2013; Friston, Kilner, & Harrison, 2006; Hohwy, 2016; Mumford, 1992).”
BackgroundRao, R. P. N., & Ballard, D. H. (1999) · Nature Neuroscience · doi:10.1038/4580 · Cited by 5,483 · cited 2×
“Before considering the implications for hierarchical architectures in the brain, we turn to the equivalent message passing for continuous variables, which transpires to be predictive coding (Rao & Ballard, 1999; Srinivasan, Laughlin, & Dubs, 1982).”
BackgroundA mechanism for cognitive dynamics: neuronal communication through neuronal coherence.
Fries, P. (2005) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2005.08.011 · Cited by 4,284 · cited 1×
“…dynamics) mechanism may be particularly important in understanding attention in terms of communication through coherence (Akam & Kullman, 2012; Fries, 2005) and the important role of inhibitory interneurons in mediating synchronous gain control (Kann, Papageorgiou, & Draguhn, 2014; Lee,…”
BackgroundHoeting, J. A., Madigan, D., Raftery, A. E., & Volinsky, C. T. (1999) · Statistical Science · doi:10.1214/ss/1009212519 · Cited by 4,118 · cited 1×
“See K. Friston and Penny (2011) and Hoeting et al. (1999) for details.”
BackgroundPreface: Cerebral Cortex Has Come of Age
Goldman-Rakic, P. S., & Rakic, P. (1991) · Cerebral Cortex · doi:10.1093/cercor/1.1.1-a · Cited by 3,393
Tracking whole-brain connectivity dynamics in the resting state
Allen, E. A., Damaraju, E., Plis, S. M., Erhardt, E. B., Eichele, T., & Calhoun, V. D. (2014) · Cerebral Cortex · doi:10.1093/cercor/bhs352 · Cited by 2,992 · cited 3×
“We focus on the modulation of intrinsic excitability; specifically, the afferents to superficial pyramidal cells—and, as a more abstract level, the implications for self-organized criticality of the sort entailed by dynamic fluctuations in connectivity (Aertsen, Gerstein, Habib, & Palm, 1989; Allen et al., 2012; Baker…”
Research works printed but not linked: 4
- Beal M. J. (2003). Variational algorithms for approximate Bayesian inference (Doctoral thesis, University College London).
- Friston K. J., Parr T., & de Vries B. (2017). Appendices for “The graphical brain: Belief propagation and active inference.” Network Neuroscience. 10.1162/netn_a_00018 PMC5798592 29417960
- Loeliger H.-A. (2002). Least squares and Kalman filtering on Forney graphs. In Blahut R. E. & Koetter R. (Eds.), Codes, graphs, and systems: A celebration of the life and career of G. David Forney, Jr., on the occasion of his sixtieth birthday (pp. 113– 135). Boston, MA: Springer US.
- MacKay D. J. C. (2003). Information theory inference and learning algorithms. Cambridge: Cambridge University Press.
Cited by
Items 101–200 of 408
Smith, R. S., Steklis, H. D., Steklis, N. G., Weihs, K. L., & Lane, R. D. (2020) · Biological Psychology · doi:10.1016/j.biopsycho.2020.107925 · Cited by 52
Smith, R. S., Moutoussis, M., & Bilek, E. (2021) · Scientific Reports · doi:10.1038/s41598-021-89047-0 · Cited by 53
“We can thus think of the mind as processing information in terms of beliefs about hidden states (e.g., see 19–21 ).”
BackgroundMethodHow Active Inference Could Help Revolutionise Robotics
Da Costa, L., Lanillos, P., Sajid, N., Friston, K., & Khan, S. (2022) · Entropy · doi:10.3390/e24030361 · Cited by 43
“, a Bayesian network), which entails the causal relationships between agent’s representations [44,45].”
BackgroundSelf-citationCognitive effort and active inference
Parr, T., Holmes, E., Friston, K. J., & Pezzulo, G. (2023) · Neuropsychologia · doi:10.1016/j.neuropsychologia.2023.108562 · Cited by 45
Self-citationA Predictive Processing Model of Perception and Action for Self-Other Distinction
Kahl, S., & Kopp, S. (2018) · Frontiers in Psychology · doi:10.3389/fpsyg.2018.02421 · Cited by 56
“The difference between continuous and discrete states in the context of active inference is well discussed in Friston et al. (2017a).”
Smith, R., Kirlic, N., Stewart, J. L., Touthang, J., Kuplicki, R., McDermott, T. J., Taylor, S., Khalsa, S. S., Paulus, M. P., & Aupperle, R. L. (2021) · Scientific Reports · doi:10.1038/s41598-021-91308-x · Cited by 44
Safron, A., Klimaj, V., & Hipólito, I. (2022) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2021.688424 · Cited by 53
“Further, while we emphasize the dynamic point of view here, we also acknowledge that brains could be understood as hybrid ‘architectures’ with both representational and non-representational sub-systems (Friston et al., 2017a; Parr and Friston, 2018).”
BackgroundActive Inference and Behavior Trees for Reactive Action Planning and Execution in Robotics
Pezzato, C., Corbato, C. H., Bonhof, S., & Wisse, M. (2023) · IEEE Transactions on Robotics · doi:10.1109/tro.2022.3226144 · Cited by 44
Time-consciousness in computational phenomenology: a temporal analysis of active inference
Bogotá, J. D., & Djebbara, Z. (2023) · Neuroscience of Consciousness · doi:10.1093/nc/niad004 · Cited by 26
“If the mapping between the parameters of active inference and neuronal structures proves to hold (Parr and Friston 2017a, 2017b, 2018a, 2018b), then not only we may begin to understand the important changes and differences of psychiatric disorders, but we may also finally begin to understand them…”
BackgroundNeural and phenotypic representation under the free-energy principle
Ramstead, M. J. D., Hesp, C., Tschantz, A., Smith, R., Constant, A., & Friston, K. (2021) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2020.11.024 · Cited by 43
Self-citationIntegrated world modeling theory expanded: Implications for the future of consciousness
Safron, A. (2022) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2022.642397 · Cited by 34
“These descending predictions are generated on multiple levels, both locally via recurrent dynamics, as well as on a more global basis, potentially accompanied by unique architectural features and discrete updating of integrative models (Friston et al., 2017b; Parr and Friston, 2018b).”
BackgroundSimulating Active Inference Processes by Message Passing
van de Laar, T. W., & de Vries, B. (2019) · Frontiers in Robotics and AI · doi:10.3389/frobt.2019.00020 · Cited by 42
“For example, when higher-order states evolve over longer timespans, hierarchical nesting leads to deep temporal models (de Vries and Friston, 2017;Friston K. et al., 2017).”
BackgroundInfluentialIlan, Y. (2022) · Computational and Structural Biotechnology Journal · doi:10.1016/j.csbj.2022.11.015 · Cited by 49
Seven computations of the social brain
Molapour, T., Hagan, C. C., Silston, B., Wu, H., Ramstead, M. J. D., Friston, K. J., & Mobbs, D. (2021) · Social Cognitive and Affective Neuroscience · doi:10.1093/scan/nsab024 · Cited by 48
“a neural network), where this message passing corresponds to neuronal communication over extrinsic (between cortical area) connections and the intrinsic connectivity of canonical microcircuits (Bastos et al., 2012; Shipp, 2016; Friston et al., 2017b).”
Self-citationToward the Next Generation of Retinal Neuroprosthesis: Visual Computation with Spikes
Yu, Z., Liu, J. K., Jia, S., Zhang, Y., Zheng, Y., Tian, Y., & Huang, T. (2020) · Engineering · doi:10.1016/j.eng.2020.02.004 · Cited by 51
Bayesian Filtering with Multiple Internal Models: Toward a Theory of Social Intelligence
Isomura, T., Parr, T., & Friston, K. (2019) · Neural Computation · doi:10.1162/neco_a_01239 · Cited by 38
Self-citationAn Investigation of the Free Energy Principle for Emotion Recognition
Demekas, D., Parr, T., & Friston, K. J. (2020) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2020.00030 · Cited by 49
Self-citationThe Thalamus as a Blackboard for Perception and Planning
Worden, R., Bennett, M. S., & Neacsu, V. (2021) · Frontiers in Behavioral Neuroscience · doi:10.3389/fnbeh.2021.633872 · Cited by 50
Predictive Neuronal Adaptation as a Basis for Consciousness
Luczak, A., & Kubo, Y. (2022) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2021.767461 · Cited by 51
The two kinds of free energy and the Bayesian revolution
Gottwald, S., & Braun, D. A. (2020) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1008420 · Cited by 54
“Therefore, saying that minimizing free energy also minimizes surprise (Parr and Friston, 2017), is generally only true in the sense that minimizing free energy minimizes an upper bound on surprise, however surprise itself is not minimized.”
BackgroundParr, T., Sajid, N., & Friston, K. J. (2020) · Entropy · doi:10.3390/e22050552 · Cited by 39
“This general formalism has been exploited in signal processing [22], Newtonian [23] and quantum dynamics [24], and neurobiology [25,26].”
MethodSelf-citationLearning Generative State Space Models for Active Inference
Çatal, O., Wauthier, S. T., Boom, C. D., Verbelen, T., & Dhoedt, B. (2020) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2020.574372 · Cited by 59
“…has been able to explain a wide array of anatomical and physiological aspects of brain systems (Angelucci et al., 2002; Bastos et al., 2012; Friston et al., 2017a) Furthermore it allows for an elegant treatment of the intricate relationship between perception and action while inherently…”
BackgroundFrom filters to fillers: an active inference approach to body image distortion in the selfie era
Tremblay, S. C., Tremblay, S. E., & Poirier, P. (2020) · AI & Society · doi:10.1007/s00146-020-01015-w · Cited by 49
A Bayesian model for chronic pain
Eckert, A.-L., Pabst, K., & Endres, D. (2022) · Frontiers in Pain Research · doi:10.3389/fpain.2022.966034 · Cited by 42
“It is generally feasible to perform inference on graphical models with continuous random variable nodes (72, 73).”
MethodActive inference leads to Bayesian neurophysiology
Isomura, T. (2021) · Neuroscience Research · doi:10.1016/j.neures.2021.12.003 · Cited by 44
“Therefore, this Review aims to introduce recent progress in theoretical neurobiology, demonstrating that a class of standard neural networks performs variational Bayesian inference under the form of a generative model (Friston, 2013; Friston, 2019; Parr et al., 2020; Isomura & Friston, 2020; Isomura et al., 2022).”
InfluentialReward maximization through discrete active inference
Da Costa, L., Sajid, N., Parr, T., Friston, K., & Smith, R. (2023) · Neural Computation · doi:10.1162/neco_a_01574 · Cited by 35
“Active inference accommodates deep hierarchical generative models combining both discrete and continuous state-spaces (Friston, Parr, et al., 2017; Friston et al., 2018; Parr et al., 2021).”
BackgroundSelf-citationWhyte, C. J. (2019) · Consciousness and Cognition · doi:10.1016/j.concog.2019.102763 · Cited by 35
Yang, Y., Zheng, Z., Liu, L., Zheng, H., Zhen, Y., Zheng, Y., Wang, X., & Tang, S. (2023) · Nature Communications · doi:10.1038/s41467-023-42053-4 · Cited by 46
“patterns allow neuronal elements to share and transmit signals through dynamical reconfiguration on multiple timescales, potentially relaxing the restriction on the material and energy cost of the structural connectome [40,56].”
Predictive Coding: Towards a Future of Deep Learning beyond Backpropagation?
Millidge, B., Salvatori, T., Song, Y., Bogacz, R., & Lukasiewicz, T. (2022) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2022/774 · Cited by 91
Hierarchical predictive coding in distributed pain circuits
Chen, Z. S. (2023) · Frontiers in Neural Circuits · doi:10.3389/fncir.2023.1073537 · Cited by 39
“Active inference employs actions to minimize PEs (Figure 2B), sharing some theoretical connections with control-as-inference and belief propagation (Friston et al., 2017; Millidge et al., 2020; Seymour and Mancini, 2020).”
BackgroundParvizi-Wayne, D., Sandved-Smith, L., Pitliya, R. J., Limanowski, J., Tufft, M. R. A., & Friston, K. J. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1354719 · Cited by 44
“” In the partially observable Markov decision process (POMDP) schema utilised in discrete state space active inference models, these beliefs are encoded within the so-called B tensor (Friston et al., 2017b; Da Costa et al., 2023).”
MethodSelf-citationActive Inference, Belief Propagation, and the Bethe Approximation
Schwöbel, S., Kiebel, S., & Marković, D. (2018) · Neural Computation · doi:10.1162/neco_a_01108 · Cited by 36
“…the derivations on the so-called Bethe approximation (Bethe, 1931, 1935), which will allow us to establish a formal link between the free energy principle and the set of update equations known as belief propagation (Friston, Parr, & de Vries, 2017; Yedidia, Freeman, & Weiss, 2005; Pearl, 1988).”
BackgroundSmith, R. S., Weihs, K. L., Alkozei, A., Killgore, W. D. S., & Lane, R. D. (2018) · Psychosomatic Medicine · doi:10.1097/psy.0000000000000661 · Cited by 39
“Project for a Spatiotemporal Neuroscience” – Brain and Psyche Share Their Topography and Dynamic
Northoff, G., & Scalabrini, A. (2021) · Frontiers in Psychology · doi:10.3389/fpsyg.2021.717402 · Cited by 37
An active inference account of protective behaviours during the COVID-19 pandemic
Bottemanne, H., & Friston, K. J. (2021) · Cognitive Affective & Behavioral Neuroscience · doi:10.3758/s13415-021-00947-0 · Cited by 33
“Based on sensory observations, the brain can update its representations, with an algorithmic process equivalent to an approximate form of Bayesian inference about the (hidden) state of its environment (Friston et al., 2017).”
BackgroundSelf-citationSafron, A., Çatal, O., & Verbelen, T. (2022) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2022.787659 · Cited by 35
“…via other subregions such as CA1, so allowing for attractors to take the form of sparsely-connected graphs—cf. hybrid continuous/discrete architectures based on Forney factor graphs and agent-designs based on independently controllable factors (Friston et al., 2017b; Thomas et al., 2017, 2018).”
BackgroundConstant, A., Hesp, C., Davey, C. G., Friston, K. J., & Badcock, P. B. (2021) · Computational Psychiatry · doi:10.5334/cpsy.70 · Cited by 38
Self-citationPredictive Coding: Towards a Future of Deep Learning beyond Backpropagation?
Millidge, B., Salvatori, T., Song, Y., Bogacz, R., & Lukasiewicz, T. (2022) · arXiv · doi:10.48550/arxiv.2202.09467 · Cited by 31
“…idea are presented in (Bogacz, 2020; Kinghorn, Millidge, & Buckley, 2021), and this mechanism is simply an example of the active inference (K. J. Friston et al., 2017; Millidge, Tschantz, Seth, & Buckley, 2020a) and control as inference (Attias, 2003; Levine, 2018; Toussaint & Storkey, 2006)…”
Federated inference and belief sharing
Friston, K. J., Parr, T., Heins, C., Constant, A., Friedman, D., Isomura, T., Fields, C., Verbelen, T., Ramstead, M., Clippinger, J., et al. (2024) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2023.105500 · Cited by 30
“…to underscore the fact that the variational message passing described above can be implemented in a neuronally plausible fashion (Friston et al., 2017a; Friston et al., 2017b; Parr and Friston, 2018), producing simulated electrophysiological responses that are similar to those observed empirically.”
BackgroundResultSelf-citation‘Consciousnessoids’: clues and insights from human cerebral organoids for the study of consciousness
Lavazza, A. (2021) · Neuroscience of Consciousness · doi:10.1093/nc/niab029 · Cited by 36
“The dual-stream information flow gated by L5p neurons is compatible with computational theories such as predictive coding based on the principle of free energy minimization (Friston et al. 2017).”
MethodFields, C., Fabrocini, F., Friston, K. J., Glazebrook, J. F., Hazan, H., Levin, M. G., & Marcianò, A. (2023) · IEEE Transactions on Molecular Biological and Multi-Scale Communications · doi:10.1109/tmbmc.2023.3272158 · Cited by 34
Self-citationThe free energy principle induces neuromorphic development
Fields, C., Friston, K. J., Glazebrook, J. F., Levin, M. G., & Marcianò, A. (2022) · Neuromorphic Computing and Engineering · doi:10.1088/2634-4386/aca7de · Cited by 31
Self-citationActive inference as a computational framework for consciousness
Vilas, M. G., Auksztulewicz, R., & Melloni, L. (2022) · Review of Philosophy and Psychology · doi:10.1007/s13164-021-00579-w · Cited by 28
A Variational Approach to Scripts
Albarracin, M., Constant, A., Friston, K. J., & Ramstead, M. J. D. (2021) · Frontiers in Psychology · doi:10.3389/fpsyg.2021.585493 · Cited by 35
“Active inference is an increasingly popular behavioral modeling framework that descends from older, closely related Bayesian approaches to the brain and behavior, such as predictive coding (Friston, 2010; Friston et al., 2017b).”
MethodSelf-citationLuu, P., Tucker, D. M., & Friston, K. J. (2023) · Cerebral Cortex · doi:10.1093/cercor/bhad458 · Cited by 27
Self-citationExpanding the Active Inference Landscape: More Intrinsic Motivations in the Perception-Action Loop
Biehl, M., Guckelsberger, C., Salge, C., Smith, S. C., & Polani, D. (2018) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2018.00045 · Cited by 37
“According to Friston et al. (2017b, Equation (A2) Appendix), the “expected free energy” is just the future conditional entropy of sensor values 6 given environment states.”
BackgroundDeep Active Inference and Scene Construction
Heins, R. C., Mirza, M. B., Parr, T., Friston, K. J., Kagan, I., & Pooresmaeili, A. (2020) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2020.509354 · Cited by 31
“The model’s beliefs about how hidden states ˜ s ( i ) cause observations ˜ o ( i ) are expressed as multidimensional arrays in the likelihood matrix A ( i ), m , where i indicates the index of the hierarchical level and m indicates a particular modality (Mirza et al., 2016; Friston et al., 2017d).”
BackgroundMethodInfluentialSelf-citationBelieving and Beliefs—Neurophysiological Underpinnings
Seitz, R. J. (2022) · Frontiers in Behavioral Neuroscience · doi:10.3389/fnbeh.2022.880504 · Cited by 34
“These processes are maintained in putative parallel cortico-subcortical loops in the brain which was taken as basis for computational modeling of belief formation (Friston et al., 2017).”
BackgroundSmith, R. S., Taylor, S., Stewart, J. L., Guinjoan, S. M., Ironside, M., Kirlić, N., Ekhtiari, H., White, E. J., Zheng, H., Kuplicki, R., et al. (2022) · Computational Psychiatry · doi:10.5334/cpsy.85 · Cited by 31
Active inference, morphogenesis, and computational psychiatry
Pio‐Lopez, L., Kuchling, F., Tung, A., Pezzulo, G., & Levin, M. G. (2022) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2022.988977 · Cited by 34
Limongi, R., Silva, A., Mackinley, M., Ford, S. D., & Palaniyappan, L. K. (2023) · Schizophrenia Bulletin · doi:10.1093/schbul/sbac125 · Cited by 30
Learning on Arbitrary Graph Topologies via Predictive Coding
Salvatori, T., Pinchetti, L., Millidge, B., Song, Y., Bao, T.-Y., Bogacz, R., & Lukasiewicz, T. (2022) · PubMed · doi:10.48550/arxiv.2201.13180 · Cited by 36
Goekoop, R., & de Kleijn, R. (2021) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2020.12.021 · Cited by 31
“To our knowledge, this is the first time that concepts from network science and graph theory are put together with current ideas on predictive coding to explain hierarchical Bayesian inference in living systems (e.g. Friston et al. (2017)).”
BackgroundRamstead, M. J. D., Albarracin, M., Kiefer, A. B., Klein, B., Fields, C., Friston, K. J., & Safron, A. (2023) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/6afs3 · Cited by 44
Self-citationHarris, D., North, J. S., & Runswick, O. R. (2022) · Psychological Research · doi:10.1007/s00426-022-01687-7 · Cited by 29
“Recent work has outlined how applying Bayesian brain perspectives (see Friston et al., 2017; Knill & Pouget, 2004; Rao & Ballard, 1999) to the study of sporting anticipation can help to explain how performers integrate and weight multiple sources of information to make predictions during…”
BackgroundMethodRelative fluency (unfelt vs felt) in active inference
Brouillet, D., & Friston, K. J. (2023) · Consciousness and Cognition · doi:10.1016/j.concog.2023.103579 · Cited by 36
“The link here rests upon the hierarchical optimisation of precision and its interpretation in terms of mental action and attentional selection (Limanowski, 2017; Limanowski, 2022; Metzinger, 2003).”
InfluentialSelf-citationHierarchical substrates of prediction in visual cortical spikingPreprint
Westerberg, J. A., Xiong, Y., Sennesh, E., Nejat, H., Ricci, D., Azmi, M., Durand, S., Hardcastle, B., Cabasco, H., Belski, H., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.10.02.616378 · Cited by 36
Friston, K. J., Costa, L. D., Tschantz, A., Kiefer, A. B., Salvatori, T., Neacsu, V., Koudahl, M., Heins, C., Sajid, N., Marković, D., et al. (2024) · Biological Psychology · doi:10.1016/j.biopsycho.2024.108891 · Cited by 46
Self-citationA Factor Graph Description of Deep Temporal Active Inference
de Vries, B., & Friston, K. J. (2017) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2017.00095 · Cited by 26
Self-citationBeliefs: A challenge in neuropsychological disorders
Seitz, R. J. (2021) · Journal of Neuropsychology · doi:10.1111/jnp.12249 · Cited by 25
Deep kinematic inference affords efficient and scalable control of bodily movements
Priorelli, M., Pezzulo, G., & Stoianov, I. P. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2309058120 · Cited by 29
The Many Roles of Precision in Action
Limanowski, J., Adams, R. A., Kilner, J. M., & Parr, T. (2024) · Entropy · doi:10.3390/e26090790 · Cited by 27
“This has recently been addressed within the active inference framework through the development of hierarchical, “mixed” generative models [32].”
InfluentialSelf-citationFitting predictive coding to the neurophysiological data
Spratling, M. W. (2019) · Brain Research · doi:10.1016/j.brainres.2019.146313 · Cited by 23
Arthur, T., & Harris, D. (2021) · Cortex · doi:10.1016/j.cortex.2021.09.017 · Cited by 28
Flexible intentions: An Active Inference theory
Priorelli, M., & Stoianov, I. P. (2023) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2023.1128694 · Cited by 28
“, by combining discrete and continuous Active Inference for planning and movement execution [84, 85, 86, 87].”
MethodMapping Husserlian Phenomenology onto Active Inference
Albarracin, M., Pitliya, R. J., Ramstead, M. J. D., & Yoshimi, J. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-28719-0_7 · Cited by 34
Hierarchical generative modelling for autonomous robots
Yuan, K., Sajid, N., Friston, K., & Li, Z. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-023-00752-z · Cited by 23
“Briefly, hierarchical generative models are a description of how sensory observations are generated, i.e., encodings of sensorimotor relationships relevant for motor control 49,50 .”
BackgroundInfluentialSelf-citationFrom pixels to planning: scale-free active inference
Friston, K. J., Heins, C., Verbelen, T., Costa, L. D., Salvatori, T., Marković, D., Tschantz, A., Koudahl, M., Buckley, C., & Parr, T. (2025) · Frontiers in Network Physiology · doi:10.3389/fnetp.2025.1521963 · Cited by 47
Self-citationActive inference, selective attention, and the cocktail party problem
Holmes, E., Parr, T., Griffiths, T. D., & Friston, K. J. (2021) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2021.09.038 · Cited by 29
Self-citationEmbodied decisions as active inference
Priorelli, M., Stoianov, I. P., & Pezzulo, G. (2025) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1013180 · Cited by 25
An Active Inference Approach to Dissecting Reasons for Nonadherence to Antidepressants
Smith, R. S., Khalsa, S. S., & Paulus, M. P. (2019) · Biological Psychiatry Cognitive Neuroscience and Neuroimaging · doi:10.1016/j.bpsc.2019.11.012 · Cited by 22
Albarracin, M., Hipólito, I., Tremblay, S. E., Fox, J. G., René, G., Friston, K. J., & Ramstead, M. J. D. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-47958-8_9 · Cited by 29
Self-citationThe computational pharmacology of oculomotion
Parr, T., & Friston, K. J. (2019) · Psychopharmacology · doi:10.1007/s00213-019-05240-0 · Cited by 23
“We then introduce a simple delay period oculomotor task—of the sort used extensively in primate electrophysiological studies (Funahashi 2015).”
MethodSelf-citationThe free energy principle: it’s not about what it takes, it’s about what took you there
Constant, A. (2021) · Biology & Philosophy · doi:10.1007/s10539-021-09787-1 · Cited by 24
Reverse-Engineering Neural Networks to Characterize Their Cost Functions
Isomura, T., & Friston, K. J. (2020) · Neural Computation · doi:10.1162/neco_a_01315 · Cited by 26
“A number of recent studies have used Markov decision process (MDP) generative models to elaborate schemes that minimize variational free energy (Friston, FitzGerald, Rigoli, Schwartenbeck, & Pezzulo, 2016, 2017; Friston, Parr, & de Vries, 2017; Fris-ton, Lin et al., 2017).”
BackgroundSelf-citationCultivating creativity: predictive brains and the enlightened room problem
Constant, A., Friston, K. J., & Clark, A. (2024) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2022.0415 · Cited by 19
Self-citationBody as First Teacher: The Role of Rhythmic Visceral Dynamics in Early Cognitive Development
Corcoran, A. W., Perrykkad, K., Feuerriegel, D., & Robinson, J. (2023) · Perspectives on Psychological Science · doi:10.1177/17456916231185343 · Cited by 21
“…opening up coherent lines of communication between disparate brain networks (Fries, 2005, 2015; Palva & Palva, 2018), paving the way for hierarchical message passing and model updating (Bastos et al., 2020; Chao et al., 2018; Friston, 2008; Friston, Parr, et al., 2017; Michalareas et al., 2016).”
Natural language syntax complies with the free-energy principle
Murphy, E., Holmes, E., & Friston, K. J. (2024) · Synthese · doi:10.1007/s11229-024-04566-3 · Cited by 29
“…neuronal dynamics have previously been cast as a gradient flow on free energy (known as variational free energy in physics, introduced in Feynman, 1972; see Hinton & Zemel, 1994); we refer the reader to Friston et al. (2017b) for a treatment of neuronal message passing and belief propagation.”
BackgroundMethodInfluentialSelf-citationA Bayesian Account of Psychopathy: A Model of Lacks Remorse and Self-Aggrandizing
Prosser, A., Friston, K. J., Bakker, N., & Parr, T. (2018) · Computational Psychiatry · doi:10.1162/cpsy_a_00016 · Cited by 23
Self-citationCosta, L. D., Sajid, N., Parr, T., Friston, K. J., & Smith, R. S. (2020) · arXiv (Cornell University) · Cited by 23
Self-citationActive inference through whiskers
Mannella, F., Maggiore, F., Baltieri, M., & Pezzulo, G. (2021) · Neural Networks · doi:10.1016/j.neunet.2021.08.037 · Cited by 25
Active inference tree search in large POMDPs
Maisto, D., Gregoretti, F., Friston, K. J., & Pezzulo, G. (2025) · Neurocomputing · doi:10.1016/j.neucom.2024.129319 · Cited by 25
Self-citationAn Overview of the Free Energy Principle and Related Research
Zhang, Z., & Xu, F. (2024) · Neural Computation · doi:10.1162/neco_a_01642 · Cited by 25
Modeling Motor Control in Continuous Time Active Inference: A Survey
Priorelli, M., Maggiore, F., Maselli, A., Donnarumma, F., Maisto, D., Mannella, F., Stoianov, I. P., & Pezzulo, G. (2023) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2023.3338491 · Cited by 21
“It can be shown that if η m = η , then L ( t ) m = 0 ; see [79] for a demonstration.”
BackgroundMethodInfluentialStructure learning enhances concept formation in synthetic Active Inference agents
Neacsu, V., Mirza, M. B., Adams, R. A., & Friston, K. J. (2022) · PLoS ONE · doi:10.1371/journal.pone.0277199 · Cited by 27
“For an extensive account of Active Inference and associated tenets, please see: Da Costa, Parr [1], Friston, FitzGerald [3], Friston, FitzGerald [4], Friston, Parr [7], Friston, Parr [18], Smith, Schwartenbeck [19].”
BackgroundSelf-citationSu, Y., MacGregor, L. J., Olasagasti, I., & Giraud, A. (2023) · PLoS Biology · doi:10.1371/journal.pbio.3002046 · Cited by 20
“To enable the information exchange between the continuous and higher discrete levels that were not accounted for in [26], we implemented the inversion of the spectral-to-continuous generative model using the “mixed model” framework in [70].”
InfluentialGenerative models for sequential dynamics in active inference
Parr, T., Friston, K. J., & Pezzulo, G. (2023) · Cognitive Neurodynamics · doi:10.1007/s11571-023-09963-x · Cited by 19
Self-citationOn the minimal theory of consciousness implicit in active inference
Whyte, C. J., Corcoran, A. W., Robinson, J., Smith, R., Moran, R. J., Parr, T., Friston, K. J., Seth, A. K., & Hohwy, J. (2026) · Physics of Life Reviews · doi:10.1016/j.plrev.2025.11.002 · Cited by 15
“The current state of the art in active inference depicts the brain as a hierarchical “mixed model” consisting of interacting generative models (Friston et al., 2017; Parr et al., 2021, 2022; Parr & Friston, 2018c).”
InfluentialSelf-citationTaylor, S., Lavalley, C. A., Hakimi, N., Stewart, J. L., Ironside, M., Zheng, H., White, E., Guinjoan, S., Paulus, M. P., & Smith, R. (2023) · Drug and Alcohol Dependence · doi:10.1016/j.drugalcdep.2023.110945 · Cited by 15
Geometric methods for sampling, optimization, inference, and adaptive agents
Barp, A., Costa, L. D., França, G., Friston, K. J., Girolami, M., Jordan, M. I., & Pavliotis, G. A. (2022) · Handbook of statistics · doi:10.1016/bs.host.2022.03.005 · Cited by 32
“Furthermore, it can be combined with discrete active inference to operate efficiently in generative models combining discrete and continuous states [287].”
BackgroundSelf-citationNested Selves: Self‐Organization and Shared Markov Blankets in Prenatal Development in Humans
Ciaunica, A., Levin, M. G., Rosas, F. E., & Friston, K. J. (2023) · Topics in Cognitive Science · doi:10.1111/tops.12717 · Cited by 24
“One potential way to address this question is in terms of Active Inference, that is, a process theory of sentient behavior (Friston et al., 2017), which is one of a class of approaches to formalize the processes by which independent subunits such as cells generate a higher-order system that has…”
BackgroundMethodSelf-citationNeuromodulatory Control and Language Recovery in Bilingual Aphasia: An Active Inference Approach
Sajid, N., Friston, K. J., Ekert, J. O., Price, C. J., & Green, D. W. (2020) · Behavioral Sciences · doi:10.3390/bs10100161 · Cited by 17
“discrete outcomes that are caused by discrete hidden states, and is described extensively in previous work [14,37].”
BackgroundSelf-citationThe neural network of sensory attenuation: A neuroimaging meta-analysis
Gu, J., Buidze, T., Zhao, K., Gläscher, J. P., & Fu, X. (2024) · Psychonomic Bulletin & Review · doi:10.3758/s13423-024-02532-1 · Cited by 14
“Friston and colleagues (Friston, 2005; Friston & Kiebel, 2009; Friston et al., 2017) outlined a general framework of brain regions potentially involved in predictive processing, encompassing primary sensory and motor cortices, motor-associated cortices, dorsomedial and ventromedial prefrontal…”
Active Inference, Bayesian Optimal Design, and Expected Utility
Sajid, N., Costa, L. D., Parr, T., & Friston, K. J. (2022) · Cambridge University Press eBooks · doi:10.1017/9781009026949.007 · Cited by 27
Self-citationManifestations, social impact, and decay of conceptual beliefs: A cultural perspective
Seitz, R. J., Paloutzian, R. F., & Angel, H. (2024) · Brain and Behavior · doi:10.1002/brb3.3470 · Cited by 16
On efficient computation in active inference
Paul, A., Sajid, N., Costa, L. D., & Razi, A. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124315 · Cited by 29
“The belief propagation scheme here is shown in the literature to have a degree of biological plausibility in the sense that it can be implemented by a local neuronal message-passing scheme [de Vries and Friston, 2017].”
MethodParvizi-Wayne, D. (2024) · Phenomenology and the Cognitive Sciences · doi:10.1007/s11097-024-10028-5 · Cited by 17
He, Y., Shao, X., Liu, C., Fan, C., Jefferies, E., Zhang, M., & Li, X. (2024) · Journal of Neuroscience · doi:10.1523/jneurosci.1404-24.2024 · Cited by 17
“…operate in a visual semantic context to e ffi ciently understand the meaning of written words, and our data support the suggestion that prediction generation and updating of an internal model are performed by the same hierarchical cortical network (K. Friston, 2005, 2010; K. J. Friston et al., 2017).”
Beyond Markov: Transformers, memory, and attention
Parr, T., Pezzulo, G., & Friston, K. J. (2025) · Cognitive Neuroscience · doi:10.1080/17588928.2025.2484485 · Cited by 18
Self-citationZeidman, P., Ashburner, J., Barnes, G. R., Kowalczyk, O. S., Lambert, C., Litvak, V., Nichols, T. E., Parr, T., Tierney, T. M., & Friston, K. J. (2025) · Cerebral Cortex · doi:10.1093/cercor/bhaf234 · Cited by 14
“For instance, a theoretical model developed to understand saccadic sampling patterns in visual neglect (Parr and Friston 2017) was used to motivate hypotheses about effective connectivity in healthy controls, subsequently assessed using DCM for MEG (Parr et al. 2019b).”
InfluentialSelf-citation
Metadata
- Authors
- Friston, K. J., Parr, T., & de Vries, B.
- Affiliations
- Wellcome Centre for Human Neuroimaging (GB)
- Eindhoven University of Technology, Department of Electrical Engineering, Eindhoven, The Netherlands
- Corresponding author
- Parr, Thomas
- Venue
- Network Neuroscience
- Publisher
- MIT Press
- Published
- 2017-12
- Volume
- 1
- Issue
- 4
- Pages
- 381-414
- ISSN
- 2472-1751 (ISSN-L)
- PMC
- PMC5798592
- OpenAlex
- W2623066203
- Cited by
- 415
- References
- 111
- Citations by year
- 2026: 49 · 2025: 39 · 2024: 41 · 2023: 61 · 2022: 51 · 2021: 56 · 2020: 55 · 2019: 32 · 2018: 25 · 2017: 6
- FWCI
- 18.51
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- Network Neuroscience (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- TU/e Research Portal (submittedVersion, other-oa)
- TU/e Research Portal (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- UCL Discovery (University College London) (submittedVersion)
- TU/e Research Portal (submittedVersion, cc-by)
- Funders
- Wellcome Trust (088130/Z/09/Z)
- Rosetrees Trust (173346)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Embodied and Extended Cognition
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
- Keywords
- active inference, belief propagation, deep generative models, factor graphs, Bayesian model averaging, thalamocortical loop, functional integration, reading, semantic representation, visual sampling
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.
A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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.