phenomenon · claim-type · cross-domain-transfer · mechanism · claim-relation · replication-status · limitation — grounded in the full text| phenomenon | automated molecular design in chemical space— “The goal of drug and material design is to identify novel molecules that have certain desirable properties.” | |
| claim-type | method— “We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation.” | |
| method — evidence | — “A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder, and a predictor.” | |
| object of study — evidence | — “To solve the inverse design challenge in chemistry, we convert molecules into continuous vector representations using neural networks.” | |
| research question — evidence | — “Discrete optimization methods have difficulty effectively searching large areas of chemical space because it is not possible to guide the search with gradients.” | |
| contribution — evidence | — “Continuous representations of molecules allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules.” | |
| transfer | from artificial intelligence (deep learning / generative models): a variational autoencoder mapping molecules (SMILES strings) into a continuous latent space for automatic chemical design— “We apply such generative models to chemical design, using a pair of deep networks trained as an autoencoder to convert molecules represented as SMILES strings into a continuous vector representation” | transfers[0] |
| transfer | from machine learning on text: the variational autoencoder with recurrent encoder and decoder, taken from Bowman et al.'s work on English sentences and applied to molecules as strings— “Variational autoencoders with recurrent neural network encoding/decoding were proposed by Bowman et al. in the context of written English sentences, and we followed their approach closely. 20” | transfers[1] |
| transfer | from generative modelling of images, speech and music: the idea that a low-dimensional continuous code supports interpolation and analogy, imported as the reason a molecular latent space should be navigable— “Such models usually also produce low-dimensional continuous representations of the data being modeled, allowing interpolation or analogical reasoning for natural images, 19 text, 20 speech, and music. 21 , 22” | transfers[2] |
| mechanism | encoding molecules into a continuous latent space so new molecules are generated by decoding/interpolating vectors (and properties optimized by moving in that space)— “by performing simple operations in the latent space, such as decoding random vectors” | mechanisms[0] |
| mechanism | an information bottleneck forces the network to keep only what is statistically salient - the compression is what makes the vector a representation rather than a copy— “This bottleneck—here the fixed-length continuous vector—induces the network to learn a compressed representation that captures the most statistically salient information in the data.” | mechanisms[1] |
| mechanism | the variational penalty fills in the space: adding noise to the encoding forces the decoder to handle points between molecules, so the space has no dead areas— “The intuition is that adding noise to the encoded molecules forces the decoder to learn how to decode a wider variety of latent points and find more robust representations.” | mechanisms[2] |
| mechanism | training the property predictor JOINTLY is what aligns the geometry with the chemistry - otherwise moving in the space would not move the property— “Second, if we develop a differentiable model that maps from molecular representations to desirable properties, we can enable the use of gradient-based optimization to make larger jumps in chemical space.” | mechanisms[3] |
| mechanism | invalid outputs are filtered rather than prevented - the decoder is allowed to be wrong and a cheminformatics toolkit discards the failures— “We employed the open source cheminformatics suite RDKit 30 to validate the chemical structures of output molecules and discard invalid ones.” | mechanisms[4] |
| builds-on | autoencoders (deep generative models, AI)— “a pair of deep networks trained as an autoencoder to convert molecules represented as SMILES strings into a continuous vector representation” | claim_relations[0] |
| builds-on | Bowman et al. (ref 20), variational autoencoders with recurrent encoder/decoder for English sentences— “Variational autoencoders with recurrent neural network encoding/decoding were proposed by Bowman et al. in the context of written English sentences, and we followed their approach closely.” | claim_relations[1] |
| contradicts | existing discrete search methods for chemical space (genetic algorithms and discrete interpolation techniques, refs 10-18)— “Discrete optimization methods have difficulty effectively searching large areas of chemical space because it is not possible to guide the search with gradients.” | claim_relations[2] |
| builds-on | Bowman et al., variational autoencoders for written English sentences— “Variational autoencoders with recurrent neural network encoding/decoding were proposed by Bowman et al. in the… | claim_relations[3] |
| contradicts | fixed-library and genetic-algorithm approaches to molecular search | claim_relations[4] |
| sense | representation - DISCRETE: the molecule as a symbol string, which a chemist can read and a machine can turn back into a structure | senses[0] |
| sense | representation - CONTINUOUS: the molecule as a real-valued vector, on which arithmetic and gradients are defined | senses[1] |
| sense | representation - LATENT: specifically the compressed code inside the bottleneck, which is a representation because it discards, not because it depicts 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. | senses[2] |
| sense | representation - as used across images, text, speech and music, where the same low-dimensional continuous code supports interpolation and analogy 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. | senses[3] |
| sense | inverse design challenge - searching chemistry by mapping molecules into continuous neural representations 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. | senses[4] |
| sense | Virtual screening - computationally assaying large libraries and experimentally testing selected leads 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. | senses[5] |
| sense | encoder network - converts each molecular string to a fixed-dimensional vector 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. | senses[6] |
| sense | decoder network - converts vectors back into molecular strings 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. | senses[7] |
| sense | property prediction task - predicts molecular properties from the latent vector during joint training 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. | senses[8] |
| sense | latent space - the encoder's vector space, functioning as a continuous molecular representation 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. | senses[9] |
| sense | dead areas - sparse latent-space regions that decode into invalid SMILES strings 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. | senses[10] |
| sense | doubly probabilistic model - stochastic in both encoding noise and decoder output sampling 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. | senses[11] |
| sense | spherical interpolation - follows a circular arc to remain in high-probability latent-space regions 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. | senses[12] |
| sense | slerp - the paper's short form for spherical interpolation 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. | senses[13] |
| sense | synthetic accessibility score - SAS, a property used to compare generated molecule sets 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. | senses[14] |
| sense | Quantitative Estimation of Drug-likeness - QED, ranging from zero to one with higher values more drug-like 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. | senses[15] |
| sense | Gaussian process model - a lightweight smooth-function model used to smooth the optimization landscape 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. | senses[16] |
| sense | joint training for property prediction - organizes latent-space molecules by property value 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. | senses[17] |
| sense | teacher forcing - feeds the previous sampled character to the GRU, improving valid outputs but complicating training 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. | senses[18] |
| control | only CANONICALISED SMILES strings used for training — held — rules out the same molecule appearing as several different strings - SMILES is not unique, so an uncanonicalised set would teach the model that one compound is many, and inflate apparent diversity in the latent space the paper is about 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. | controls[0] |
| edge | information bottleneck —part-of→ autoencoder 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. | concept_edges[1] |
| edge | SMILES —contrasted-with→ InChI 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. | concept_edges[2] |
| audit | information bottleneck —causes→ latent representation — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but endpoint by anaphora: 'This bottleneck' = the information bottleneck (previous sentence); 'latent representation' is written 'compressed representation'. 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. | concept_edges[4] |
| note | encoder network —part-of→ autoencoder — The paper explicitly says the autoencoder is comprised of the encoder and decoder 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. | concept_edges[6] |
| note | decoder network —part-of→ autoencoder — The paper explicitly says the autoencoder is comprised of the encoder and decoder 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. | concept_edges[7] |
| outcome (stated by the tagger, no quotation) | control — held | controls[0] |
| for which use | limitation — any claim of coverage over chemical space: OPTIMIZATION IN MOLECULAR SPACE IS EXTREMELY CHALLENGING, BECAUSE THE SEARCH SPACE IS LARGE, DISCRETE, AND UNSTRUCTURED 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. | limitations[0] |
| for which use | limitation — the generated molecules: THE AUTOENCODER SOMETIMES PRODUCED MOLECULES THAT ARE FORMALLY VALID AS GRAPHS BUT CONTAIN MOIETIES THAT ARE NOT DESIRABLE BECAUSE OF STABILITY OR SYNTHETIC CONSTRAINTS 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. | limitations[1] |
| for which use | limitation — the decoder's output: THE CHARACTER-BY-CHARACTER NATURE OF THE SMILES REPRESENTATION AND THE FRAGILITY OF ITS INTERNAL SYNTAX CAN STILL RESULT IN THE OUTPUT OF INVALID strings 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. | limitations[2] |
| for which use | limitation — the learning problem as posed: FORCING THE DECODER TO PRODUCE VALID SMILES STRINGS MAKES THE LEARNING PROBLEM UNNECESSARILY HARD SINCE THE DECODER MUST ALSO IMPLICITLY LEARN WHICH STRINGS ARE VALID SMILES 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. | limitations[3] |
| for which use | limitation — any claim about generating novel molecules: THE PROBABILITY OF DECODING FROM A POINT IN LATENT SPACE IS DEPENDENT ON HOW CLOSE THIS POINT IS TO THE LATENT REPRESENTATIONS OF OTHER MOLECULES - a decoding rate of 73-79% NEAR known molecules 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. | limitations[4] |
| for which use | limitation — the property-prediction comparison: PERFORMANCE FOR ELECTRONIC PROPERTIES ARE SIMILAR TO GRAPH CONVOLUTIONS FOR SOME PROPERTIES, and accuracy COULD BE IMPROVED 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. | limitations[5] |
| for which use | limitation — the graph-output alternative the authors prefer: BUILDING A NEURAL NETWORK WHICH CAN OUTPUT ARBITRARY GRAPHS IS AN OPEN PROBLEM 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. | limitations[6] |
| duplicate of | limitation — limitations[2] 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. | limitations[7] |
| for which use | limitation — the same scope as limitations[2] - SMILES syntax fragility still yields invalid output 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. | limitations[7] |
| for which use | limitation — the latent space without the variational constraint: IT MAY BE SPARSE AND MAY CONTAIN LARGE DEAD AREAS, WHICH DECODE TO INVALID SMILES STRINGS 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. | limitations[8] |
| for which use | limitation — every result in the paper, which is conditional on the encoding: WE ALSO TESTED InChI AS AN ALTERNATIVE STRING REPRESENTATION, BUT FOUND IT TO PERFORM SUBSTANTIALLY WORSE THAN SMILES, PRESUMABLY DUE TO A MORE COMPLEX SYNTAX THAT INCLUDES COUNTING AND ARITHMETIC 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. | limitations[9] |
| replication | untested | |
| no type for this | spherical interpolation → slerp — The parenthetical introduces slerp as the short name for spherical interpolation. This is an abbreviation/short-form relation: it is not class membership, containment, identity of two independently named concepts, distinction, contrast, operationalisation, causation, or renaming across fields. 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. | unhoused_relations[0] |
| limitation | optimization in molecular space is hard because the search space is large, discrete, and unstructured 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. | limitations[0] |
| limitation | Decoded molecules can be graph-valid but chemically undesirable. 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. | limitations[1] |
| limitation | SMILES syntax fragility means the decoder still emits invalid molecules despite the variational constraint. 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. | limitations[2] |
| limitation | Requiring the decoder to learn SMILES validity makes the learning problem harder than it needs to be. 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. | limitations[3] |
| limitation | Decoding succeeds only near known molecules; random latent points almost always fail. 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. | limitations[4] |
| limitation | Property-prediction accuracy is only comparable to graph convolutions for some properties and is not fully optimised. 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. | limitations[5] |
| limitation | The graph-output alternative the authors prefer is not yet solvable. 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. | limitations[6] |
| limitation | The decoder still emits chemically invalid strings, because SMILES syntax is fragile and the model works character by character. 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. | limitations[7] |
| limitation | Without the variational constraint the latent space contains dead regions that decode to nothing valid. 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. | limitations[8] |
| limitation | The choice of string encoding is load-bearing and only partly explained - InChI performed substantially worse and the reason is offered as a presumption. 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. | limitations[9] |
finding · quantity · variable · definition · key-term — grounded in the full text| finding | An autoencoder maps SMILES-encoded molecules to a continuous vector space. 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.findings[0] |
| finding | EVIDENCE CORRECTED (RE-GROUNDED) · Continuous representations allow gradient-based optimization to guide the search for optimized functional compounds. 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.findings[1] |
| retracted | finding — RE-GROUNDED. The published string was in NEITHER copy of the paper. Ours read 'THESE representations'; the paper says 'CONTINUOUS representations', and continuity is the property the whole method rests on - it is what makes gradient-based optimization possible at all. The demonstrative dropped the reason. | content_tags.findings[1] |
| finding | Once molecules are points in a continuous space, generation becomes arithmetic - decode a random vector, perturb a known molecule, interpolate between two. 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.findings[2] |
| finding | The encoding matters: SMILES works and InChI works substantially worse, apparently because its syntax involves counting and arithmetic. 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.findings[3] |
| finding | The search problem is one of scale: about 10^8 substances have ever been made, against an estimated 10^23 to 10^60 drug-like possibilities. 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.findings[4] |
| finding | Genetic generation of compounds requires manually specified heuristics for mutation and crossover. 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.findings[5] |
| quantity | hundreds of thousands chemical structures — training set size of existing molecules 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.quantities[0] |
| quantity | < 9 heavy atoms — size regime of one demonstration molecule set 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.quantities[1] |
| quantity | 10^8 synthesised vs 10^23–10^60 possible — the size of the search problem the method addresses 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.quantities[2] |
| variable | continuous representation (independent) 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.variables[0] |
| variable | property values (dependent) 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.variables[1] |
| variable | string representation (independent) 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.variables[2] |
| definition | SMILES strings 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[1] |
| definition | latent representation 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 | information bottleneck 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] |
| definition | variational autoencoder 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[4] |
| definition | encoder 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[5] |
| definition | decoder 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[6] |
| definition | predictor 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[7] |
| key-terms | molecules, continuous, representation, encoder, decoder, chemical, latent, optimization, SMILES, latent space, autoencoder, variational autoencoder, InChI, information bottleneck, chemical space, gradient-based optimization, inverse design |
| concept | variational autoencoder 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[1] |
| concept | latent representation — not defined in this paper's own words; the fixed-length continuous vector a molecule is encoded into - a compressed, machine-manipulable stand-in for the molecule 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[4] |
| concept | SMILES — not defined in this paper's own words; a text encoding of an organic molecule, chosen because a string can be turned straight back into a molecule 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 | InChI — not defined in this paper's own words; the rival string encoding, tested and found substantially worse for this purpose 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 | autoencoder — not defined in this paper's own words; a pair of networks trained to reproduce their input through a bottleneck, so that the bottleneck learns a compressed code 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 | information bottleneck — not defined in this paper's own words; the narrow layer that forces compression, and thereby forces the network to keep what matters 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 | chemical space — not defined in this paper's own words; the space of possible molecules - large, discrete and unstructured, which is why searching it is hard 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] |
| concept | gradient-based optimization — not defined in this paper's own words; the search method the whole exercise is designed to make available, by giving chemical space a differentiable coordinate system 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[10] |
| concept | inverse design challenge — not defined in this paper's own words; the chemistry problem addressed by converting molecules into continuous vector representations for search and optimization 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[11] |
| concept | Virtual screening — not defined in this paper's own words; assaying large virtual libraries with simulations or learned proxy models and experimentally testing only promising leads 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[12] |
| concept | encoder network — not defined in this paper's own words; the autoencoder component that converts each molecular string into a fixed-dimensional vector 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[13] |
| concept | decoder network — not defined in this paper's own words; the autoencoder component that converts fixed-dimensional vectors back into molecular strings 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[14] |
| concept | property prediction task — not defined in this paper's own words; the joint training objective in which a multilayer perceptron predicts a molecule property from its latent vector 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[15] |
| concept | latent space — not defined in this paper's own words; the vector space produced by the encoder, functioning as a continuous molecular representation 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[16] |
| concept | dead areas — not defined in this paper's own words; sparse latent-space regions whose points decode to invalid SMILES strings 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[17] |
| concept | doubly probabilistic model — not defined in this paper's own words; a model stochastic both in the noisy encoding and in sampling the decoder's character outputs 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[18] |
| concept | spherical interpolation — not defined in this paper's own words; interpolation that keeps samples in high-probability regions by following a circular arc on an N-dimensional sphere 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[19] |
| concept | slerp — not defined in this paper's own words; the short name used for spherical interpolation along an arc on an N-dimensional sphere 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[20] |
| concept | synthetic accessibility score — not defined in this paper's own words; SAS, one of the chemical properties used to compare generated molecules 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[21] |
| concept | Quantitative Estimation of Drug-likeness — not defined in this paper's own words; QED, a zero-to-one measure whose higher values indicate a more drug-like molecule 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[22] |
| concept | Gaussian process model — not defined in this paper's own words; a lightweight smooth-function model used to create a smoother property landscape for latent-space optimization 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[23] |
| concept | joint training for property prediction — not defined in this paper's own words; training that organizes the latent-space distribution of molecules by property values 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[24] |
| concept | teacher forcing — not defined in this paper's own words; training the GRU decoder with the previously sampled character as an additional input, improving valid-string accuracy while increasing training difficulty 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[25] |
| Bayesian | 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 | Bayesian — 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] |
| mean μ | as written: mean 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[1] |
| regression β | as written: regression model 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[2] |
| standard-deviation SD/σ | 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[3] |
| not machine-vouched | standard-deviation — 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[3] |
| regression MLP predictor | 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[4] |
| method | computational-model — a second reading DIFFERING from the recorded value; recorded: system-or-tool-building 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. | alternatives.method |
| phenomenon | the discreteness of chemical space as an obstacle to searching it — a second reading DIFFERING from the recorded value; recorded: automated molecular design in chemical space 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. | alternatives.phenomenon |
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