phenomenon · claim-type · cross-domain-transfer · mechanism · claim-relation · replication-status · limitation — grounded in the full text| phenomenon | molecular property prediction from graphs— “To investigate the success of MPNNs on predicting chemical properties, we use the publicly available QM9 dataset (Ramakrishnan et al., 2014).” | |
| claim-type | method— “In this paper, we reformulate existing models into a single common framework we call Message Passing Neural Networks (MPNNs) and explore additional novel variations within this framework.” | |
| method — evidence | — “We performed numerous experiments in order to find the best possible MPNN on this dataset as well as the proper input representation.” | |
| transfer | from artificial intelligence (deep learning): graph neural networks / deep learning, applied to predict quantum-chemical properties of molecules— “using machine learning (and especially deep learning)” | transfers[0] |
| mechanism | message passing on the molecular graph (atoms as nodes, bonds as edges): iteratively passing and aggregating messages between atoms— “a message passing algorithm and aggregation procedure to compute a function of their entire input graph” | mechanisms[0] |
| builds-on | graph neural networks (invariant to molecular symmetries)— “neural network models invariant to molecular symmetries have already been described in the literature” | claim_relations[0] |
| builds-on | Gated Graph Neural Networks (GG-NN), Li et al. (2016)— “We began our exploration of MPNNs around the GG-NN model which we believe to be a strong baseline.” | claim_relations[1] |
| builds-on | the set2set model, Vinyals et al. (2015)— “Second is a set2set model from Vinyals et al. (2015).” | claim_relations[2] |
| extends | Interaction Networks, Battaglia et al. (2016)— “Thus we also tried using a variant on the message function as described in (Battaglia et al., 2016).” | claim_relations[3] |
| sense | 'chemical accuracy' - TWO REFERENCE POINTS, AND THE PAPER PICKS THE NEARER ONE: the chemistry community's target error is relative to nature, DFT error is the gap between the simulation and nature, and this paper states that its own use of the phrase means error against its available labels - so the headline measures agreement with a simulation, not with a molecule— “The first is the estimated average error of the DFT approximation to nature, which we refer to as “DFT error.” The second, known as “chemical accuracy,” is a target error that has been established by the chemistry community” | senses[0] |
| sense | MPNN - A NAME GIVEN TO EIGHT EXISTING MODELS AFTER THE FACT: the framework is introduced as a reformulation of models already in the literature under one label, which is the inverse of the usual naming failure - here a paper renames its field rather than being renamed by it— “we reformulate existing models into a single common framework we call Message Passing Neural Networks (MPNNs) and explore additional novel variations within this framework” | senses[1] |
| sense | the prediction target - A MODEL OF A MODEL: what the network predicts is the outcome of a computationally expensive DFT calculation, so the object being learned is itself a simulation of the quantity of interest rather than the quantity— “A Message Passing Neural Network predicts quantum properties of an organic molecule by modeling a computationally expensive DFT calculation” | senses[2] |
| control | the model family evaluated on all 13 targets of the public QM9 benchmark — held — rules out a chemistry result earned on molecules of the authors' own choosing - QM9 is fixed, public and shared with the methods being compared against— “We develop an MPNN which achieves state of the art results on all 13 targets and predicts DFT to within chemical accuracy on 11 out of 13 targets.” | controls[0] |
| edge | chemical accuracy —operationalised-as→ ground truth labels— “In the rest of this paper when we talk about chemical accuracy we generally mean with respect to our available ground truth labels” | concept_edges[0] |
| edge | chemical accuracy —distinct-from→ DFT error— “The first is the estimated average error of the DFT approximation to nature, which we refer to as “DFT error.” The second, known as “chemical accuracy,” is a target error that has been established by the chemistry community” | concept_edges[1] |
| outcome (stated by the tagger, no quotation) | control — held | controls[0] |
| for which use | limitation — every accuracy claim: the target is THE DFT APPROXIMATION, so 'chemical accuracy' is accuracy to a model rather than to the quantum properties themselves— “the ability to fit the DFT approximation to within chemical accuracy would be an encouraging step” | limitations[0] |
| for which use | limitation — the scalability claim: THE TOWERS VARIATION MAKES THESE MODELS MORE SCALABLE, BUT ADDITIONAL IMPROVEMENTS WILL BE NEEDED TO SCALE TO MUCH LARGER GRAPHS— “The towers variation makes these models more scalable, but additional improvements will be needed to scale to much larger graphs.” | limitations[1] |
| for which use | limitation — the spatial-information variants specifically: GENERALIZING TO LARGER MOLECULE SIZES SEEMS PARTICULARLY CHALLENGING WHEN USING SPATIAL INFORMATION— “Generalizing to larger molecule sizes seems particularly challenging when using spatial information.” | limitations[2] |
| for which use | limitation — the towers-plus-edge-network combination: UNFORTUNATELY, OUR ATTEMPTS SO FAR AT COMBINING THE TOWERS AND EDGE NETWORK MESSAGE FUNCTION HAVE FAILED TO FURTHER IMPROVE PERFORMANCE— “Unfortunately, our attempts so far at combining the towers and edge network message function have failed to further improve performance, possibly because the combination makes training more difficult.” | limitations[3] |
| for which use | limitation — the ablations using partial charge, which no deployment could reproduce: THIS FEATURE IS AN OUTPUT OF THE DFT CALCULATION AND THUS COULD NOT BE USED IN AN APPLIED SETTING— “This feature is an output of the DFT calculation and thus could not be used in an applied setting.” | limitations[4] |
| for which use | limitation — every accuracy figure on this benchmark, which is agreement with a computation rather than with measurement: THE DATASET CONTAINING THE TRUE TARGETS ON ALL 134k MOLECULES DOES NOT CURRENTLY EXIST— “The dataset containing the true targets on all 134k molecules does not currently exist.” | limitations[5] |
| for which use | limitation — the claim that this model reaches chemical accuracy: the authors state that in the rest of the paper the phrase means accuracy with respect to their available ground truth labels, not the community's target error relative to nature— “In the rest of this paper when we talk about chemical accuracy we generally mean with respect to our available ground truth labels” | limitations[6] |
| for which use | limitation — a comparison of the errors reported here against either standard: the paper states that for all 13 targets, achieving chemical accuracy is at least as hard as achieving DFT error— “For all 13 targets, achieving chemical accuracy is at least as hard as achieving DFT error” | limitations[7] |
| for which use | limitation — the claim that this framework unifies prior work: at least eight models from the literature are described in it, and the exact expressions and the derivation of that reformulation are in the supplementary material rather than the paper— “There are at least eight notable examples of models from the literature that we can describe using our Message Passing Neural Networks” | limitations[8] |
| replication | untested | |
| limitation | the models fit the DFT approximation, itself an approximation of true quantum properties— “the ability to fit the DFT approximation to within chemical accuracy would be an encouraging step” | limitations[0] |
| limitation | The towers variant improves scalability but is still insufficient for much larger graphs.— “The towers variation makes these models more scalable, but additional improvements will be needed to scale to much larger graphs.” | limitations[1] |
| limitation | Generalising to larger molecules is hard precisely when spatial information is used.— “Generalizing to larger molecule sizes seems particularly challenging when using spatial information.” | limitations[2] |
| limitation | Combining the towers variant with the edge-network message function did not work.— “Unfortunately, our attempts so far at combining the towers and edge network message function have failed to further improve performance, possibly because the combination makes training more difficult.” | limitations[3] |
| limitation | Some ablation experiments used a partial-charge input feature that would not be available in a real application.— “This feature is an output of the DFT calculation and thus could not be used in an applied setting.” | limitations[4] |
| limitation | Accuracy is measured against DFT approximations because experimentally true targets do not exist for the full dataset.— “The dataset containing the true targets on all 134k molecules does not currently exist.” | limitations[5] |
| limitation | The paper's headline term is redefined to mean error against its own labels rather than against nature, and it says so in one sentence.— “In the rest of this paper when we talk about chemical accuracy we generally mean with respect to our available ground truth labels” | limitations[6] |
| limitation | Two different error standards are in play and the paper orders them.— “For all 13 targets, achieving chemical accuracy is at least as hard as achieving DFT error” | limitations[7] |
| limitation | Eight existing models are re-described in the authors' own framework, with the derivations placed outside the paper.— “There are at least eight notable examples of models from the literature that we can describe using our Message Passing Neural Networks” | limitations[8] |
finding · quantity · variable · definition · key-term — grounded in the full text| finding | MPNN variants achieve a new state of the art on the QM9 dataset— “we are able to both achieve a new state of the art on the QM9 dataset” | content_tags.findings[0] |
| finding | neural inference is far cheaper than the DFT simulation it approximates— “the inference time of the neural networks discussed in this work is 300k times faster” | content_tags.findings[1] |
| finding | the readout must be permutation-invariant for graph-isomorphism invariance— “R operates on the set of node states and must be invariant to permutations of the node states” | content_tags.findings[2] |
| finding | Duvenaud-style message passing cannot capture edge-node correlations— “is unable to identify correlations between edge states and node states” | content_tags.findings[3] |
| quantity | 130k molecules — size of the QM9 benchmark with 13 properties each— “QM9 consists of 130k molecules with 13 properties for each molecule which are approximated by an expensive1 quantum mechanical simulation method (DFT), to yield 13… | content_tags.quantities[0] |
| quantity | 300k times faster — MPNN inference speedup over DFT simulation | content_tags.quantities[1] |
| quantity | 11 of 13 targets — targets predicted within chemical accuracy by the best MPNN | content_tags.quantities[2] |
| quantity | 5 of 13 targets — targets hit using molecular topology alone | content_tags.quantities[3] |
| evidence status note | quantity — marked quote-not-in-current-staged-text - today's merge gate would REFUSE this quote: Paper.locate() does not find it in the current staged text, and no disclosure on the row explains why. Update, 15 September 2026: CLEARED evidence status='quote-not-in-current-staged-text' - the quote was replaced (EXTEND) and the new quote locates exactly once in the current staged text. | content_tags.quantities[3] |
| quantity | 8 models — prior models unified under the MPNN framework | content_tags.quantities[4] |
| variable | hidden states (per-node states updated during message passing) | content_tags.variables[1] |
| variable | time steps (number of message-passing rounds) | content_tags.variables[2] |
| variable | node features (graph inputs to the network) | content_tags.variables[4] |
| variable | edge features (graph inputs to the network) | content_tags.variables[5] |
| variable | update functions (learned differentiable components of the MPNN) | content_tags.variables[6] |
| variable | readout functions (learned differentiable components of the MPNN) | content_tags.variables[7] |
| definition | Message Passing Neural Networks (MPNNs) | content_tags.definitions[0] |
| definition | chemical accuracy | content_tags.definitions[1] |
| definition | DFT error Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[2] |
| definition | readout phase Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[3] |
| key-terms | message passing, quantum chemistry, molecular property, graph isomorphism, MPNN, DFT, chemical accuracy, readout function, node features, hidden states, supervised learning |
| concept | chemical accuracy Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[5] |
| concept | DFT error Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[6] |
| concept | Message Passing Neural Networks Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[7] |
| concept | ground truth labels Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[8] |
| concept | QM9 Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[9] |
| mean μ | as written: average of Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[0] |
| not machine-vouched | mean — not-in-our-copy - The quote is in the original web page we fetched for this paper, but not in the text we tag against, which was extracted differently. Checked with ligatures, hyphens, spacing and punctuation set aside. | content_tags.statistics[0] |
| sample-size n/N | as written: N Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[1] |
When we have changed our mind. Some entries in the table begin with a word in capitals and the reason for it. EVIDENCE CORRECTED — the finding still stands, but we had quoted the wrong sentence from the paper, and the quotation has been replaced with the right one. RETRACTED — the entry itself was wrong; it stays on the page, marked, with the reason, rather than being deleted. WITHDRAWN AND REPLACED — the entry was wrong, and a corrected version of it appears in the entries below. Each mark carries its own reason, in brackets, because the reasons differ from one entry to the next. An entry we have WITHDRAWN is removed from this page and kept in the record, with its reason. A line reading Audit and a date is a later re-reading of the entry against its whole sentence: DOUBTFUL means it was kept, and the reason says what is loose about it. Quotation marks inside the tagger's sentence. Where a tag is the tagger's own sentence, a phrase in quotation marks inside it is borrowed from the paper — the tagger's wording around it is not. The paper's own sentence is the evidence quote for that row; on this page it is withheld with the article, and the article is at its publisher. Nothing inside quotation marks anywhere on this page is ours: if it is quoted, it is the paper's. Whose words a concept carries. A concept entry records a term the paper itself uses, so the term, the sense given for it, and its evidence are the paper's own words, quoted where the paper fixes them. This is the opposite of a sense entry in the tag table, which is the tagger's sentence about how the paper uses a term. Where a concept entry could not be grounded in the paper's words, the entry says so on its own row.
Tag layer © the Meta-Tagging Project, CC BY 4.0. The paper text is © its authors/publisher
and is linked, not reproduced. Excerpts shown are short quotations for identification and verification.
· back to index
· concepts · authors
· עברית
© 2026 Shir Sivroni · ORCID 0009-0000-2597-2824