Molecular Property Prediction

Follow this topic Ask about this topic

Momentum

19 papers in the last four weeks, up 46% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 133

May 28, 2026cs.LG

MōLe-Λ: Learning the Coupled-Cluster Response State for Energies, Gradients, and Properties

Coupled-cluster (CC) theory is often considered the gold standard of quantum chemistry, but its high computational cost limits routine access to accurate energies, forces and response properties. While the right-hand

T

-amplitudes determine the correlated wavefunction, many practically important observables additionally require the left-hand

Λ

-amplitudes. We introduce MōLe-

Λ

, an extension of Molecular Orbital Learning (MōLe) that predicts the full ground-state coupled-cluster singles and doubles (CCSD) response state by jointly learning right-hand amplitudes

(T_1,T_2)

and left-hand amplitudes

(Λ_1,Λ_2)

from localized Hartree--Fock molecular orbitals. Architecturally, MōLe-

Λ

extends MōLe with

Λ_1

and

Λ_2

readouts that mirror the symmetry constraints of the

T_1

and

T_2

heads, while preserving the original equivariant orbital encoder, odd sign-equivariant decoding, locality and size-extensivity. The resulting model yields accurate CC-quality energies and forces, while simultaneously recovering dipoles, quadrupoles, polarizabilities, the electron density, and 2-electron observables such as the pair density. We show that MōLe-

Λ

further extends the speed advantage of MōLe over full CCSD while substantially expanding the accessible properties, providing a route to wavefunction-level surrogate models for correlated quantum chemistry.

Andreas Burger, Luca Thiede, Abdulrahman Aldossary +4

University of Toronto, Canada · Vector Institute for Artificial Intelligence, Toronto, Canada · NVIDIA, Toronto, Canada +2

Neural Surrogate Modeling Quantum Chemistry Molecular Property Prediction

May 26, 2026cs.LG

SCENT: Aligning Mass Spectra with Molecular Structure for Olfactory Perception

Predicting human olfactory perception from molecular structure has seen remarkable progress, yet these approaches require explicit chemical structure at inference, which is not available in practical sensing settings. We address this gap by exploring direct electron ionization mass spectrometry (EI-MS), a sensing technique that acquires chemically informative fragmentation fingerprints in seconds, as an alternative input modality for olfactory prediction. We contribute Spectrum-to-Chemical Embedding alignmeNT (SCENT), a multi-modal contrastive learning framework that aligns EI-MS representations with pretrained chemical structure embeddings, while requiring only mass spectra at inference. On the multi-label odor descriptor prediction task, SCENT significantly outperforms MS-only baselines and achieves performance comparable to structure-based models, despite requiring no explicit molecular structure at test time. The learned representations also better approximate continuous human perceptual ratings and generalize to real-world lab-measured spectra, suggesting that cross-modal alignment is an effective strategy for grounding analytical spectra in chemical semantics.

Ziqi Zhang, Eunyeong Jin, Miguel Vasco +6

1Dept. of Intelligent Systems, KTH Royal Institute of Technology, Stockholm, Sweden · 2Atmospheric Chemistry Dept., Max Planck Institute for Chemistry, Mainz, Germany · 3Dept. of Information Technology, Uppsala University, Uppsala, Sweden +1

Cross-Modal Alignment Molecular Property Prediction Mass Spectrometry Molecular Representation Learning

May 25, 2026q-bio.QM

What Molecular Structure Cannot Tell Us: A Taxonomy of Explainability Gaps in GNN-Based Drug Toxicity Prediction

Not all clinically relevant adverse effects are structurally inferable from molecular graphs - regardless of model quality or architectural complexity. This study introduces an operational taxonomy of the structural information limits that prevent structure-based toxicity prediction, independent of the learning algorithm employed. Graph Neural Networks (GNNs) have emerged as a natural approach for molecular toxicity prediction, operating directly on atomic connectivity without the information loss inherent to fixed-length fingerprints. However, the fraction of a drug's known pharmacological profile that is actually inferable from molecular structure remains systematically underexplored. A systematic case study using acetylsalicylic acid (ASA, Aspirin) - one of the most comprehensively characterized drugs in pharmacology - serves as model compound. A Message Passing Neural Network (MPNN) is trained on the Tox21 benchmark and GNNExplainer is applied to characterize atom-level attribution. Results indicate that molecular structure explains approximately 45% (5/11) of known ASA adverse effects. A four-category Gap Taxonomy (GAP-1 through GAP-4) is introduced distinguishing between principally non-encodable effects, data gaps arising from Missing Not At Random (MNAR) mechanisms, assay panel mismatches, and representation errors. The MNAR gap is empirically quantified via a systematic ChEMBL query (42 documented assays, 0 retrievable bioactivity entries). An attention pooling experiment localizes the representation error to the MPNN message passing layers rather than the aggregation step. The Gap Taxonomy has direct implications for drug safety signal detection and regulatory frameworks including Good Pharmacovigilance Practice (GVP) guidelines and New Approach Methodologies (NAMs). Structural limits identified are confirmed in a companion DDI ablation study.

Juergen Dietrich

Molecular Property Prediction GNN Explainability

May 23, 2026cs.LG

Aligning Molecular Graph Explanations with Chemical Identity via InChIfied Invariants

Obtaining consistent explanations for machine learning on molecular graphs requires predictions and attributions to be aligned with chemical identity. However, chemically equivalent drawings of the same molecule can induce different molecular representations, leading to inconsistent predictions and explanations. Here, we introduce InChIfied Invariants, a class of node, edge, and graph features based on the International Chemical Identifier (InChI) and designed to be invariant under transformations that preserve chemical identity. Using one million molecular graphs from PubChem Substances, we show that InChIfied Invariants produce identical representations for chemically equivalent graphs in 99.62% of cases, whereas standard Daylight invariants do so in only 0.35% of cases. Across MoleculeNet tasks, InChIfied Invariants preserve predictive performance while significantly improving prediction consistency across alternative graph depictions of the same molecules. We further perform a quantitative attribution analysis and show that explanations produced with standard molecular featurization methods vary substantially across chemically equivalent graphs, while InChIfied Invariants enforce consistent attributions by construction. We release open-source software implementing InChIfied Invariants, which can be used as a drop-in replacement for standard molecular graph features.

Emanuele Guidotti, Sara Puglioli

University of Lugano, Switzerland · Philochem AG, Switzerland

Molecular Property Prediction Invariant Representation Learning GNN Explainability Molecular Representation Learning

May 19, 2026cs.LG

Hierarchical Contrastive Learning for Multi-Domain Protein-Ligand Binding

Predicting protein-ligand binding affinity remains intractable for multi-domain proteins, where inter-domain dynamics govern molecular recognition. Existing geometric deep learning methods typically treat proteins as monolithic static graphs, suffering from rigid-body assumptions and aleatoric noise in flexible regions. To address this, we introduced HCLBind, a self-supervised framework that decouples geometric representation learning from affinity regression. HCLBind leverages a general-to-specific pre-training paradigm on the Q-BioLiP database to learn a robust physical grammar of binding. We propose a novel hierarchical decoy strategy: the model learns local physicochemical constraints through protein coordinate perturbation in single-domain proteins and global conformational geometry through inter-domain rotation in multi-domain complexes. Our hybrid architecture integrates a domain-gated graph attention network and cross-modal attention to explicitly prioritize domain interfaces. Furthermore, we employ LoRA on protein and ligand foundation models, ensuring efficient optimization while preserving evolutionary knowledge. Experiments on PDBBind demonstrate that HCLBind effectively learns discriminative interface features and provides robust uncertainty estimation, overcoming the limitations of standard supervised learning. The code is available at https://github.com/jiankliu/HCLBind.

Shuo Zhang, Rongqi Hong, Huifeng Zhang +1

University of Birmingham, UK

Protein-Ligand Binding Affinity Prediction Contrastive Learning Self-Supervised Pre-Training Molecular Property Prediction

May 17, 2026cs.LG

When Molecular Similarity Works: Property Cliffs Reveal Hidden Errors

Accurate prediction of molecular properties underpins drug discovery and material design, yet even state-of-the-art models remain vulnerable to localized failure modes that aggregate metrics cannot detect. The places where molecular similarity should be most helpful are also places where standard evaluation can be most misleading. Property cliffs expose this gap: structurally similar molecules can still differ sharply in target property, so models with competitive overall performance may fail in high-risk local neighborhoods. To expose and mitigate this failure mode, CliffSplit, a cliff-aware evaluation protocol that constructs locally supported, cliff-exposed test cases, and CliffLoss, a model-agnostic train-only mitigation mechanism for cliff-sensitive errors, are introduced. Experiments on three QM9 targets and three MoleculeNet tasks across five backbones show that CliffSplit reveals at least 15% higher error in cliff-heavy QM9 regions, while CliffLoss reduces the cliff-to-smooth error gap by up to 30% on Lipophilicity and improves overall MAE by 9.7%. Together, these results turn molecular similarity failure from a descriptive anomaly into a benchmarked evaluation problem for molecular machine learning. The code is available at https://anonymous.4open.science/r/Cliff_Loss.

Di Hu, Kun Li, Haojie Rao +6

School of Economics and Management · School of Computer Science, Wuhan University Wuhan, China · Department of Data Science and Artificial Intelligence, Monash University Victoria, Australia +2

Molecular Property Prediction

May 16, 2026cs.LG

Tensor Channel Equivariant Graph Neural Networks for Molecular Polarizability Prediction

We introduce a tensor-channel equivariant graph neural network for direct prediction of molecular polarizability tensors. Building on the efficient PaiNN architecture, we augment the hidden representation with explicit symmetric rank-2 tensor channels aligned with the decomposition of polarizability into isotropic and anisotropic components. In contrast to approaches that construct tensor outputs only at readout, our model propagates tensor structure throughout message passing using geometrically motivated tensor bases. This yields a target-aligned architecture for tensor-valued molecular prediction. On optimized QM7-X geometries, the proposed model achieves lower full-tensor and anisotropic error than both a PaiNN-style readout baseline and a dielectric MACE baseline under matched training conditions and at nearly identical parameter count. In this controlled setting, it also outperforms MACE while remaining substantially faster at inference. Ablation studies show that the gain does not arise from increased capacity alone, but from the combination of explicit tensor propagation and a traceless target parameterization matched to the anisotropic part of the polarizability tensor. Among the tensor bases considered, the strongest results are obtained from interactions between learned directional features, indicating that these are particularly effective for modeling molecular polarizability. Rotational equivariance tests further confirm that all compared models are numerically equivariant, so the observed improvements are attributable to better learning of the target tensor itself. Overall, our results show that for structured tensor-valued targets, propagating target-aligned tensor features can outperform both readout-only tensor construction and a more general higher-order equivariant model in the present training setting.

Jean Philip Filling, Daniel Franzen, Michael Wand

Institute for Computer Science, Johannes Gutenberg University Mainz, Germany

Equivariant GNNs Molecular Property Prediction

May 15, 2026cs.LG

Structure-Aware Masking for Protein Representation Learning

Masked language modeling (MLM) is the standard objective for training protein language models, typically implemented by randomly masking individual residues at a fixed rate (e.g., 15%). This practice implicitly assumes that all sequence positions contribute equally to representation learning. In downstream fitness prediction tasks, however, protein sequences are governed by three-dimensional structural dependencies and long-range residue contacts that induce strong nonlocal couplings between residues. We introduce Bucket Masking, a structure-aware masking strategy that selects groups of residues based on their proximity in three-dimensional space, preferentially masking structurally coupled regions during training. By conditioning the masking distribution on residue contacts, Bucket Masking shifts the learning objective toward modeling long-range interactions that are critical for protein function. Across four downstream protein fitness prediction tasks, Bucket Masking enables up to a 14% improvement over standard random masking, excelling at predicting higher-order mutational interactions. Through controlled ablations, we show that these improvements arise from mask placement rather than span size, establishing masking as a positional inductive bias.

Thomas Walton, Ayan Goel, Amirali Aghazadeh

Georgia Institute of Technology

Protein Fitness Prediction Masked Language Modeling Protein Language Models Molecular Property Prediction Protein Representation Learning

May 15, 2026cs.LG

Multi-level Self-supervised Pretraining on Compositional Hierarchical Graph for Molecular Property Prediction

Self-supervised pretraining on molecular graphs has emerged as a promising approach for molecular property prediction, yet most existing methods operate at a single structural granularity and treat bond information as auxiliary edge attributes rather than as an independent semantic layer. In this work, we propose MolCHG, a multi-level self-supervised pretraining framework built upon a novel Compositional Hierarchical Graph that organizes molecular structure into four types of nodes across three semantic levels. By introducing a bond graph that operates in parallel with the atom graph, our architecture elevates bond-level information to independently evolving node representations, enabling fragment nodes to aggregate atom-level and bond-level semantics on an equal footing. We design three level-specific pretraining objectives: an atom-bond cross-view contrastive task that aligns the atom-view and bond-view representations within each fragment, a fragment-level functional group prediction task to inject domain-relevant chemical knowledge, and graph-level structure prediction tasks to encode global molecular topology. Experiments on nine MoleculeNet benchmarks demonstrate that MolCHG achieves the best performance on seven datasets across both classification and regression tasks, remaining competitive with the strongest baselines on the rest. Ablation studies further confirm that the multi-level supervision signals are complementary and that each component contributes to the overall performance.

Xiayu Liu, Zhengyi Lu, Hou-biao Li

School of Mathematical Sciences, University of Electronic Science and Technology of China, No.2006, Xiyuan Avenue, West Hi-tech Zone,2026 611731, Chengdu, China · Department of Computer Science and Engineering, Oakland University, 201 Meadow Brook RD, Rochester Hills, 48309, MI, USA

Self-Supervised Pre-Training Hierarchical Representation Learning Molecular Property Prediction Molecular Representation Learning

May 14, 2026cs.LG

DrugSAGE:Self-evolving Agent Experience for Efficient State-of-the-Art Drug Discovery

Building state-of-the-art (SOTA) predictive models for drug discovery requires expensive search over tools, architectures, and training strategies. Current LLM-based agents can find SOTA solutions through extensive trial and error, but they do not retain the experience accumulated along the way and therefore pay the full search cost on every new task. We propose \method (Self-evolving Agent Experience), a framework that accumulates and reuses experience across tasks to build SOTA drug discovery models efficiently. \method maintains a cross-task memory of verified skills, statistical evidence about effective strategies, and a record of recurring errors and their fixes. In some cases, \method transfers a working solution directly without test-time search. In 33 molecular property prediction tasks, \method ranks first among nine SOTA agents in a single-task setting. With memory accumulated from 16 smaller tasks, \method achieves an averaged normalized score of 0.935 on 17 held-out tasks in a cross-task evaluation setting and outperforms all baseline agents by 10-30% in a zero-test-time search regime. In summary, our work shows the advantage of cross-task memory for efficient SOTA model development in drug discovery.

Yikun Zhang, Xiwei Cheng, Tianyu Liu +2

1Northeastern University · 2Broad Institute of MIT and Harvard · 3Yale University +1

AI Agents for Scientific Discovery AI-Assisted Scientific Research Molecular Property Prediction Memory-Augmented Agents Self-Evolving Agents Drug Discovery

May 13, 2026cs.LG

Learning to Select Source Domains: Proxy-Rewarded Policy Optimization for Molecular OOD Generalization

Molecular property prediction under severe out-of-distribution (OOD) shifts remains challenging because conventional scaffold splits can retain local structural similarity between training and test molecules, while adaptation from heterogeneous source domains may cause negative transfer. We introduce SCOPE-Bench, a scaffold-cluster benchmark constructed by partitioning Bemis-Murcko scaffolds in an explicit physicochemical descriptor space, and POMA, a retrieve-compose-adapt framework for source-domain selection when target property labels are unavailable. POMA uses unlabeled target structures to retrieve labeled source scaffolds as proxy targets. A policy is trained from reductions in proxy-task mean absolute error after adaptation, and the selected source subset is then used for target adaptation with covariance alignment at the whole-molecule and BRICS-derived substructure levels. We evaluate HOMO, LUMO, and HOMO-LUMO gap prediction on QM9 using three 3D molecular backbones and 15 target-scaffold tasks. Relative to conventional scaffold splitting, prediction errors on SCOPE-Bench increase by up to 8.0-fold, with a mean increase of 5.9-fold. Relative to the supervised baseline under the strict OOD split, POMA reduces mean absolute error by up to 11.2%, with an average relative reduction of 6.2% over the nine backbone-property combinations. These results support adaptive source selection as a useful strategy for molecular prediction under strong structural distribution shifts.

Zhuohao Lin, Kun Li, Jiameng Chen +4

School of Computer Science Wuhan University Wuhan, China · College of Computer Science and Technology, Zhejiang University Hangzhou, China · School of Life Sciences and Technology, Tongji University Shanghai, 200092, China +1

Unsupervised Domain Adaptation Domain Adaptation Domain Generalization OOD Generalization Molecular Property Prediction

May 13, 2026cs.LG

Chem-GMNet: A Sphere-Native Geometric Transformer for Molecular Property Prediction

Modern SMILES-based chemical language models obtain strong MoleculeNet performance by treating SMILES as generic text and compensating with multi-million-molecule self-supervised pretraining. We ask: when a domain carries structural priors as rich as chemistry's, does it warrant a domain-native transformer rather than a generic one rescued by scale? We answer affirmatively with \textbf{GM-Net} (Geometric Measure Network), a transformer family in which every module is replaced by a sphere-native counterpart, and instantiate it as \textbf{Chem-GMNet}. Three blocks follow: SH-Embedding (tokens as learnable directions on

S^{k-1}

lifted through a Gegenbauer feature map); DualSKA (a per-head fusion of a linear-time gated Sphere-Flow recurrence whose persistent state we prove is the truncated multipole expansion of the input distribution, and a softmax Sphere-Kernel branch over the same Schoenberg-valid kernel); and SH-FFN (sphere projection

\to

Gegenbauer lift

\to

moment readout). On canonical DeepChem scaffold splits, against same-shape ChemBERTa-2 baselines under the chemberta3-faithful protocol: (i) random-initialised, Chem-GMNet wins on 7 of 10 MoleculeNet endpoints at

\sim\!35\%

fewer parameters; (ii) pretrained on the same 10M-SMILES ZINC corpus as ChemBERTa-2 MLM-10M, it matches or beats the public release on 6 of 8 shared endpoints (5/7 excluding a known ClinTox release anomaly). A

(k,L)

ablation shows that increasing the sphere dimension from

k\!=\!8

to

k\!=\!10

at fixed

L\!=\!3

lowers ESOL RMSE to

0.938

at scratch, beating pretrained ChemBERTa-2 MLM-10M on this endpoint without any pretraining at all.

Deepak Warrier, Raja Sekhar Pappala

MSTACK AI

Transformer Geometric Representation Learning Molecular Property Prediction Molecular Representation Learning

May 12, 2026cs.LG

Decomposing the Generalization Gap in PROTAC Activity Prediction: Variance Attribution and the Inter-Laboratory Ceiling

Machine-learning predictors of biochemical activity often exhibit large random-split-to-leave-one-target-out generalisation gaps that have been documented but not decomposed. We frame this as an evaluation-science question and use targeted protein degradation as the empirical test bed. PROTACs (proteolysis-targeting chimeras) are heterobifunctional small molecules that induce targeted protein degradation, with more than forty candidates currently in clinical trials; published predictors report AUROC of 0.85 to 0.91 under random-split cross-validation, while the leave-one-target-out (LOTO) protocol of Ribes et al. reduces performance to approximately 0.67. Random splits reward within-target interpolation, whereas LOTO measures the novel-target prediction that de-novo design depends on. We decompose this gap and identify inter-laboratory measurement variance as the dominant component, anchored by a within-target cross-laboratory cascade bounding the inter-laboratory contribution at 0.124 AUROC, well above the 0.05 contribution from binarisation-threshold choice. Across eight published architectures and ESM-2 protein language models up to 3B parameters, LOTO AUROC plateaus near 0.67, with a comparable plateau under SMILES-level deduplication; a 21-dimensional 2000-trial hyperparameter optimisation cannot break this ceiling, and the rank-1 single-seed configuration regresses by 0.161 AUROC under multi-seed validation, matching a closed-form selection-bias prediction (Bailey and Lopez de Prado, 2014). Few-shot k=5 stratified per-target retraining combined with ADMET features lifts 65-target LOTO AUROC from 0.668 to 0.7050, and post-hoc Platt scaling recovers raw output to within the 0.05 well-calibrated threshold. We release PROTAC-Bench (10,748 measurements, 173 targets, 65 LOTO folds), the variance-decomposition framework, the per-target calibration protocol, and the evaluation code.

Thor Klamt, Wolfgang Nejdl, Ming Tang

L3S Research Center, Leibniz Universität Hannover, Appelstraße 9a, 30167 Hannover, Germany · Institute of Data Science (Knowledge-Based Systems), Faculty of Electrical Engineering and Computer Science,2026 Leibniz Universität Hannover, Appelstraße 9a, 30167 Hannover, Germany

Molecular Property Prediction Scientific Reproducibility Drug Discovery

May 11, 2026cs.LG

On Improving Graph Neural Networks for QSAR by Pre-training on Extended-Connectivity Fingerprints

Molecular Graph Neural Networks (GNNs) are increasingly common in drug discovery, particularly for Quantitative Structure-Activity Relationship (QSAR) studies; yet, their superiority compared to classical molecular featurisation approaches is disputed. We report a general strategy for improving GNNs for QSAR by pre-training to predict Extended-Connectivity Fingerprints (ECFP). We validate our approach with statistical tests and challenging out-of-distribution (OOD) splits. Across five out of six Biogen benchmarks, we observed a statistically significant improvement in standard performance metrics over all evaluated baselines when using ECFP pre-trained GNNs. However, for more heterogeneous datasets and more complex endpoints, such as binding affinity prediction, pre-trained GNNs underperformed in OOD settings. Importantly, we investigated the impact of substructure-level data leakage during pre-training on downstream performance. While we identified scenarios where pre-training on ECFPs was less effective, our findings show that ECFP-based pre-training can enhance downstream OOD performance on a diverse set of practically relevant QSAR tasks.

Sam Money-Kyrle, Markus Dablander, Thierry Hanser +3

Department of Statistics, University of Oxford, 24-29 St Giles’, Oxford, OX1 3LB, United Kingdom. · Mathematical Institute, University of Oxford, Andrew Wiles Building, Woodstock Road, Oxford, OX2 6GG, United Kingdom. · Molecular Informatics and AI, Lhasa Limited, Granary Wharf House,May 2 Canal Wharf, Leeds, LS11 5PS, United Kingdom.

Graph Neural Networks OOD Generalization Molecular Property Prediction Molecular Representation Learning

May 11, 2026cs.LG

It's All Connected: Topology-Aware Structural Graph Encoding Improves Performance on Polymer Prediction

Graph Neural Networks (GNNs) have achieved strong results in molecular property prediction, but polymers present distinct challenges: labeled datasets are scarce and small (typically in the order of hundreds of polymers) due to the need for expensive experimentation, and complex polymer chain distributions influence polymer properties. Established practice in polymer prediction represents polymers solely by graphs of their repeat units, discarding the chain-scale morphology that governs key properties such as the glass transition temperature (

T_g

). In this work, we propose a principled graph construction that addresses this gap. Given a polymer's molecular mass distribution (MMD), we sample representative chains from the Schulz-Zimm distribution and construct representative sets of large graphs encoding chain-scale topology directly, with atoms and bonds featurized using rich chemical descriptors. We further pretrain GNN encoders via masked graph modeling on 100,000 unlabeled PSMILES strings before fine-tuning on labeled data. On a dataset of 381 polymers (180 homopolymers and 201 copolymers), we show that graph construction and self-supervised pretraining are jointly necessary: without pretraining, the large graph method matches the repeat-unit baseline (28.40 K vs. 28.36 K RMSE); with pretraining, it achieves 24.76 K +/- 3.30 K, a 5.1% reduction in mean error over the pretrained repeat-unit baseline (26.08 K +/- 4.20 K, p < 0.001, 30 runs). An ablation removing chemical features degrades performance to 36.65 K, confirming both components are essential. Results are architecture-agnostic, holding for both GINE and GATv2 encoders.

H. Ibrahim Erdogan, Punith Raviswamy, Nikita Agrawal +5

University of Bayreuth, Germany · Faculty of Engineering Science, University of Bayreuth, Germany · Faculty of Mathematics, Physics & Computer Science, University of Bayreuth, Germany +5

Graph Representation Learning Materials Property Prediction Molecular Property Prediction Molecular Representation Learning

May 11, 2026physics.chem-ph

Physical probes expose and alleviate chemical-environment collapse in molecular representations

Nuclear magnetic resonance (NMR) spectroscopy provides an experimental readout of local chemical environments, but its use in molecular representation learning has been constrained by heterogeneous data and incomplete atom-level assignments. Here we construct complementary high-fidelity experimental and computational 13C NMR resources, which reveal a recurrent form of representational collapse: atoms that are equivalent in molecular topology can remain experimentally distinct in their real chemical environments, whereas explicit 3D descriptions are further limited by static conformations in dynamic regimes. To alleviate this bottleneck, we develop CLAIM (Contrastive Learning for Atom-to-molecule Inference of Molecular NMR), a framework that aligns efficient topological molecular inputs with atom-resolved NMR observables. Through hierarchical chemical priors and cross-level contrastive learning, CLAIM restores lost chemical resolution and markedly improves atom-level molecule-spectrum retrieval. CLAIM remains robust in flexible and tautomeric systems for 13C NMR prediction, improves stereoisomer discrimination without explicit 3D modelling, and transfers to broader molecular property tasks including ADMET prediction and fluorescence estimation. These results establish physically grounded spectral alignment as an effective strategy for alleviating chemical-environment collapse and for guiding experimentally grounded molecular representation learning.

Jiebin Fang, Zidi Yan, Churu Mao +7

Hainan Institute, Zhejiang University, Sanya 572025, China · Institute of Marine Biology and Pharmacology, Ocean College, Zhejiang University, Zhoushan 316021, China · School of Food and Pharmacy, Zhejiang Ocean University, Zhoushan 316021, China +2

Contrastive Learning Cross-Modal Representation Learning Molecular Property Prediction Molecular Representation Learning

May 11, 2026cs.CV

MolSight: Molecular Property Prediction with Images

Every molecule ever synthesised can be drawn as a 2D skeletal diagram, yet in modern property prediction this universally available representation has received less focus in favour of molecular graphs, 3D conformers, or billion-parameter language models, each imposing its own computational and data-engineering overhead. We present

\textbf{MolSight}

, the first systematic large-scale study of vision-based Molecular Property Prediction (MPP). Using 10 vision architectures, 7 pre-training strategies, and

2\,M

molecule images, we evaluate performance across 10 downstream tasks spanning physical-property regression, drug-discovery classification, and quantum-chemistry prediction. To account for the wide variation in structural complexity across pre-training molecules, we further propose a

\textbf{chemistry-informed curriculum}

: five structural complexity descriptors partition the corpus into five tiers of increasing chemical difficulty, consistently outperforming non-curriculum baselines. We show that a single rendered bond-line image, processed by a vision encoder, is sufficient for competitive molecular property prediction, i.e.

\textit{chemical insight from sight alone}

. The best curriculum-trained configuration achieves the top result on

\textbf{5 of 10}

benchmarks and top two on

\textbf{all 10}

, at

\textbf{

\textit{80

\times

lower}

}

FLOPs than the nearest multi-modal competitor.

Aaditya Baranwal, Akshaj Gupta, Yogesh S Rawat +1

University of Central Florida, Orlando, Florida, USA · Birla Institute of Technology and Science, Pilani, Rajasthan, India

Curriculum Learning Molecular Property Prediction Molecular Representation Learning

May 9, 2026cs.LG

Benchmarking Compositional Generalisation for Machine Learning Interatomic Potentials

Machine Learning Interatomic Potentials play a fundamental role in computational chemistry and materials science, enabling applications from molecular dynamics simulations to drug design and materials discovery. While recent approaches can estimate inter-atomic forces with high precision, it remains unclear to what extent they can generalise to previously unseen molecules. Do they learn the compositional structure of chemistry, capturing how molecular fragments and their combinations determine properties, or do they primarily learn to interpolate patterns that are specific to the training examples? To address this question, we propose a benchmark consisting of four tasks that require some form of compositional generalisation. In each task, models are tested on molecules that were unseen during training, but the training data is chosen such that generalisation to the test examples should be feasible for models that learn the underlying physical principles. Our empirical analysis shows that the considered tasks are highly challenging for state-of-the-art models, with errors on out-of-distribution examples often an order of magnitude higher than on in-distribution examples, even when using foundation models that have been pre-trained on millions of molecules.

Amir Masoud Nourollah, Irtaza Khalid, Stefano Leoni +1

Cardiff University Cardiff, CF10 3AT, UK

Machine Learning Interatomic Potentials OOD Generalization Molecular Property Prediction Compositional Generalization

May 9, 2026cs.LG

Structural Interpretations of Protein Language Model Representations via Differentiable Graph Partitioning

Protein language models such as ESM-2 learn rich residue representations that achieve strong performance on protein function prediction, but their features remain difficult to interpret as structural

\&

evolutionary signals are encoded in dense latent spaces. We propose a plug-

\&

-play framework that projects ESM-2 representations onto protein contact graphs

\&

applies

\textbf{SoftBlobGIN}

, a lightweight Graph Isomorphism Network with differentiable Gumbel-softmax substructure pooling, to perform structure-aware message passing

\&

learn coarse functional substructures for downstream prediction tasks. Across enzyme classification, SoftBlobGIN achieves 92.8% accuracy

\&

0.898 macro-F1. Unlike post hoc analysis of protein language models alone, our method produces directly auditable structural explanations: GNNExplainer recovers biologically meaningful active-site residues, spatially localized functional clusters,

\&

catalytic contact patterns. On binding-site detection, SoftBlobGIN improves residue AUROC from

0.885

using an ESM-2 linear probe to

0.983

, indicating that these structural explanations are not recoverable from language-model features alone. Learned blob partitions provide an additional layer of interpretability by automatically grouping residues into functional substructures, with blobs containing annotated active-site residues showing

1.85\times

higher importance than other blobs (

ρ{=}0.339

,

p{=}0.009

), without any active-site supervision. Our framework requires no retraining of the language model, adds only

\sim

1.1M parameters,

\&

generalises across ProteinShake tasks, achieving

F_{\max}

of

0.733

on Gene Ontology prediction

\&

AUROC of

0.969

on binding-site detection. We position this as an interpretable structural companion to protein language models that makes their predictions more transparent

\&

auditable.

Siddhant Dutta, Edward Tan Beng Wai, Soumick Sarker +2

Nanyang Technological University, Singapore

Graph Neural Networks Protein Language Models Molecular Property Prediction Protein Representation Learning GNN Explainability

May 7, 2026cs.LG

Better Protein Function Prediction by Modeling Survivorship Bias

Protein sequence data from nature exhibits survivorship bias: we only observe data from those organisms that survive and reproduce, while non-functional protein mutations are eliminated by natural selection. Thus, predicting whether a protein sequence is functional often requires learning from positive examples alone. While positive-unlabeled (PU) learning frameworks offer a generic solution to this problem, existing PU methods ignore the evolutionary processes that shape sequence observability and cause survivorship bias. Consider a sequence that is one mutation away from a commonly-observed protein variant in a well-surveilled organism. If the sequence were functional, it would likely be observed. If it is not observed, this suggests non-functionality. In contrast, sequences that are unlikely to arise through mutation may be missing simply because they never arose. Thus, these two kinds of missing sequences should be treated differently when training models. In this work, we propose Evo-PU, a PU learning framework that uses a scientific understanding of nucleotide mutation to model survivorship bias for well-surveilled single-organism sequence data. On three prediction tasks using single-organism uniform-coverage surveillance data -- predicting results from held-out influenza and respiratory syncytial virus (RSV) mutagenesis studies, and predicting future SARS-CoV-2 variants -- Evo-PU outperforms standard PU learning, one-class classification (OCC), and protein language models (PLMs). On prediction tasks from multi-organism ProteinGym datasets with more heterogeneous surveillance coverage, we identify opportunities to generalize our approach.

Zhongmou Chao, Poompol Buathong, Ekaterina Selivanovitch +2

Smith School of Chemical and Biomolecular Engineering, Cornell University, USA · Center for Applied Mathematics, Cornell University, USA · School of Operations Research and Information Engineering, Cornell University, USA

Weakly Supervised Learning Molecular Property Prediction Sequence Modeling Positive-Unlabeled Learning

May 7, 2026cs.LG

Edge-specific signal propagation on mature chromophore-region 3D mechanism graphs for fluorescent protein quantum-yield prediction

Fluorescent protein quantum yield (QY) is governed by the mature chromophore and its three-dimensional microenvironment rather than sequence identity alone. Protein language models and emission-band averages capture global trends, but do not model how local physical signals act on specific chromophore regions. We present a chromophore-centred mechanism graph algorithm for QY prediction. Each PDB structure is converted into a typed 3D residue graph, registered to a mature-CRO state, partitioned into phenolate, bridge and imidazolinone regions, and transformed by channel-signal-region propagation. The representation contains 121 enrichment features; after removing identity shortcuts, 52 non-identity features are used for band-specific ExtraTrees regression. Because each feature encodes a contact channel, seed signal and target CRO region, interpretation is intrinsic rather than post hoc. On a 531-protein benchmark, the method achieved the best random-CV performance among model-based baselines (R = 0.772 +/- 0.008, MAE = 0.131 +/- 0.002), exceeding Band mean (R = 0.632), ESM-C (R = 0.734) and SaProt (R = 0.731), and ranked first in bright screening (Bright P@5 = 0.704). Under homology control, the advantage was clearest in the remote bucket (<50% similarity; R = 0.697 versus 0.633, 0.575 and 0.408), with the strongest overall bright/dark Top-K screening. Stable selected features recovered band-specific mechanisms: aromatic packing and clamp asymmetry in GFP-like proteins, charge/clamp balance in Red proteins, and flexibility-risk/bulky-contact features in Far-red proteins. Source code, feature tables and evaluation scripts are available from the first author upon request. Contact: [email protected]

Yuchen Xiong, Swee Keong Yeap, Steven Aw Yoong Kit

1China-ASEAN College of Marine Sciences, Xiamen University Malaysia, Sepang 43900, Selangor, Malaysia

Graph Representation Learning Molecular Property Prediction

May 4, 2026cs.LG

Bolek: A Multimodal Language Model for Molecular Reasoning

Molecular property models increasingly support high-stakes drug-discovery decisions, but their outputs are often difficult to audit: classical predictors return scores without rationale, while language models can produce fluent explanations weakly grounded in the input molecule. We introduce Bolek, a compact multimodal language model that grounds natural-language reasoning in molecular structure by injecting a Morgan fingerprint embedding into an instruction-tuned text decoder. Bolek is fine-tuned on molecular alignment tasks, including molecule description, RDKit descriptor prediction, and substructure detection, and on downstream reasoning over 15 TDC binary classification tasks using synthetic chains-of-thought anchored in concrete molecular features. Across these tasks, Bolek outperforms its Qwen3-4B-Instruct base on all endpoints in yes/no mode and on 13 of 15 in chain-of-thought mode, raising mean ROC/PR AUC from 0.55 to 0.76. It also outperforms TxGemma-9B-Chat on 13 of 15 binary classification tasks despite being less than half its size. Bolek's explanations are more grounded than those of the baseline LLMs: it cites numerical descriptors 10-100x more often per chain-of-thought, and the cited values agree strongly with RDKit for key descriptors such as TPSA, MolLogP, and MolWt (Spearman rho = 0.87-0.91). Generalisation extends beyond the training panel: on 15 unseen TDC classification endpoints, Bolek matches TxGemma on five, and it produces non-trivial rank correlations on three held-out regression endpoints despite never seeing downstream regression during training. These results suggest that targeted modality injection and reasoning supervision tied to verifiable molecular features can yield compact, auditable molecular reasoning models.

Frederic Grabowski, Jacek Szczerbiński, Maciej Jaśkowski +4

1Ingenix.ai, Warsaw, Poland · Laboratory of Bioinformatics and Computational Genomics, Warsaw University of Technology, Warsaw, Poland

LLM Grounding Scientific Reasoning Multimodal Large Language Models Molecular Property Prediction Molecular Representation Learning

May 1, 2026cs.LG

A Comparative Study of QSPR Methods on a Unique Multitask PAMPA dataset

We present a unique, multitask dataset comprising 143 drug and drug candidate molecules, each evaluated on in vitro, parallel artificial-membrane permeability assays (PAMPA) using six different model membranes. Using this resource, we systematically assess the effectiveness of various molecular descriptors and regression models in predicting passive membrane permeability. The studied models range from simple linear regression to a modern pre-trained transformer architecture. Particular attention is given to the trade-off between predictive performance and model interpretability, highlighting the challenges introduced by machine learning approaches. To our knowledge, this is the most comprehensive study on simultaneous modeling of multiple organ-specific PAMPA membranes to date, offering novel insights into membrane-specific permeability profiles. We found that expert-designed physico-chemical property descriptors are more fitting for a limited sample size permeabilty study than deep learning based representations.

Andrs Formanek, Anna Vincze, Richrd Bicsak +3

Department of Electrical Engineering (ESAT), STADIUS Center for Dynamical Systems,May Signal Processing and Data Analytics, KU Leuven, 3001 Leuven, Belgium · Department of Artificial Intelligence and Systems Engineering, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Műegyetem rkp. 3, H-1111 Budapest, Hungary[cs.LG] · Department of Chemical and Environmental Process Engineering, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3, H-1111 Budapest, Hungary +2

Multi-Task Learning Molecular Property Prediction Molecular Representation Learning

Apr 30, 2026cs.LG

Hyper-Dimensional Fingerprints as Molecular Representations

Computational molecular representations underpin virtual screening, property prediction, and materials discovery. Conventional fingerprints are efficient and deterministic but lose structural information through hash-based compression, particularly at low dimensionalities. Learned representations from graph neural networks recover this expressiveness but require task-specific training and substantial computational resources. Here we introduce hyperdimensional fingerprints (HDF), which replace the learned transformations of message-passing neural networks with algebraic operations on high-dimensional vectors, producing deterministic molecular representations without any training. Across diverse property prediction benchmarks, HDF outperforms conventional fingerprints in the majority of tasks while exhibiting greater consistency across datasets and models. Crucially, HDF embeddings preserve molecular similarity faithfully: at 32 dimensions, distances in HDF space achieve a 0.9 Pearson correlation with graph edit distance, compared to 0.55 for Morgan fingerprints at equivalent size. This structural fidelity persists at low dimensions where hash-based methods degrade, allowing simple nearest-neighbor regression to remain predictive with as few as 64 components. We further demonstrate the practical impact in Bayesian molecular optimization, where HDF-based surrogate models achieve substantially improved sample efficiency in regimes where Morgan fingerprints perform comparably to random search. HDF thus provides a general-purpose, training-free alternative to conventional molecular fingerprints, suggesting that the information loss long accepted as inherent to fixed-length fingerprints is a limitation of the hash-based encoding scheme rather than the fingerprint paradigm itself.

Jonas Teufel, Luca Torresi, André Eberhard +1

Karlsruhe Institute of Technology (KIT), Institute of Nanotechnology (INT), Kaiserstraße 12, Karlsruhe, Germany. · Karlsruhe Institute of Technology (KIT), Institute of Anthropomatics and Robotics (IAR), Kaiserstraße 12, Karlsruhe, Germany.

Hyperdimensional Computing Molecular Property Prediction Molecular Representation Learning

Apr 29, 2026cs.LG

Do Larger Models Really Win in Drug Discovery? A Benchmark Assessment of Model Scaling in AI-Driven Molecular Property and Activity Prediction

The rapid growth of molecular foundation models and large language models (LLMs) has encouraged a scale centred view of AI in drug discovery, in which larger pretrained models are expected to supersede compact cheminformatics models. We test this assumption across 26 ADME, toxicity and bioactivity endpoints, covering 165,541 endpoint level compound label records. The benchmark contains 78 endpoint and split entries evaluated under random, Murcko scaffold and structure separated 5-fold cross validation protocols, representing increasing chemical generalization difficulty. Across 156 task and metric comparisons, classical machine learning (ML) provides the largest share of best performing entries (47.4%), followed by pretrained molecular sequence models (28.8%), graph neural networks (21.8%) and LLM based SAR baselines (1.9%). Classical ML dominates random split interpolation and remains the largest winner family overall. GNN and sequence models are competitive in selected harder splits, but their strict winner shares decrease under a fixed final-window readout, indicating sensitivity to training settings and model selection. Paired bootstrap analyses show that small numerical differences between individual models should not be read as decisive victories. SAR knowledge from training folds improves GPT5.5-SAR and Opus4.7-SAR metrics but does not make rule based reasoning a universal substitute for supervised predictors. Compact specialized models remain highly effective, and predictive performance depends on the fit among model, task and validation scenario, not on scale alone.

Jinjiang Guo, Sheng Ding

Global Health Drug Discovery Institute, Beijing 100084, China · School of Pharmaceutical Sciences, Tsinghua University, Beijing 100084, China

LLM Evaluation Molecular Property Prediction Molecular Representation Learning Drug Discovery

Apr 27, 2026physics.chem-ph

Enhancing molecular dynamics with equivariant machine-learned densities

Machine-learning interatomic potentials (MLIPs) have enabled molecular dynamics at near ab initio accuracy, yet remain limited to energies and forces by construction, leaving electronic observables such as dipole moments and polarizabilities inaccessible. We introduce DenSNet, a density-first approach to machine-learned electronic structure that learns the Hohenberg--Kohn map from nuclear configurations to the ground-state electron density. Our approach employs an SE(3)-equivariant neural network to predict density coefficients of a flexible atom-centered Gaussian basis, combined with a

Δ

-learning strategy that uses superposed atomic densities as a prior to accelerate training. A second equivariant network then maps the predicted density to the total energy, providing a unified framework for molecular dynamics and electronic structure. We validate DenSNet on ethanol, ethanethiol, and resorcinol, where infrared spectra from machine-learned trajectories show excellent agreement with experimental gas-phase measurements. To test scalability, we train on polythiophene oligomers with 1--6 monomers and extrapolate to chains of up to 12 monomers, generating stable long-time trajectories whose infrared spectra agree with reference density functional theory calculations. Here, we show that reinstating the electron density as the central learned quantity opens a practical route to transferable prediction of spectroscopic and electronic observables in large-scale molecular simulations.

Mihail Bogojeski, Muhammad R. Hasyim, Leslie Vogt-Maranto +3

BIFOLD and Machine Learning Group, Technische Universität Berlin, Franklinstr. 28/29, 10587 Berlin, Germany · Department of Chemistry, New York University, New York, NY 10003, USA · Department of Artificial Intelligence, Korea University, Seoul 02841, Korea +5

Machine Learning Interatomic Potentials Quantum Chemistry Molecular Property Prediction Molecular Dynamics Simulation

Apr 25, 2026cs.LG

h-MINT: Modeling Pocket-Ligand Binding with Hierarchical Molecular Interaction Network

Accurate molecular representations are critical for drug discovery, and a central challenge lies in capturing the chemical environment of molecular fragments, as key interactions, such as H-bond and π stacking, occur only under specific local conditions. Most existing approaches represent molecules as atom-level graphs; however, atom-level representations can hardly express higher-order chemical context (e.g., stereochemistry, lone pairs, conjugation). Fragment-based methods (e.g., principal subgraph, predefined functional groups) fail to preserve essential information such as chirality, aromaticity, and ionic states. This work addresses these limitations from two aspects. (i) OverlapBPE tokenization. We propose a novel data-driven molecule tokenization method. Unlike existing approaches, our method allows overlapping fragments, reflecting the inherently fuzzy boundaries of small-molecule substructures and, together with enriched chemical information at the token level, thereby preserving a more complete chemical context. (ii) h-MINT model. OverlapBPE induces many-to-many atom-fragment mappings, which necessitate a new hierarchical architecture. We therefore develop a hierarchical molecular interaction network capable of jointly modeling interactions at both atom and fragment levels. By supporting fragment overlaps, the model naturally accommodates the many-to-many atom-fragment mappings introduced by the OverlapBPE scheme. Extensive evaluation against state-of-the-art methods shows our method improves binding affinity prediction by 2-4% Pearson/Spearman correlation on PDBBind and LBA, enhances virtual screening by 1-3% in key metrics on DUD-E and LIT-PCBA, and achieves the best overall HTS performance on PubChem assays. Further analysis demonstrates that our method effectively captures interactive information while maintaining good generalization.

Yanru Qu, Yijie Zhang, Wenjuan Tan +6

Department of Computer Science, UIUC · School of Computer Science, McGill University · MILA-Québec AI Institute +4

Protein-Ligand Binding Affinity Prediction Hierarchical Representation Learning AI-Assisted Scientific Research Molecular Property Prediction Molecular Representation Learning Drug Discovery

Apr 25, 2026cs.LG

HBGSA: Hydrogen Bond Graph with Self-Attention for Drug-Target Binding Affinity Prediction

Accurate prediction of drug-target binding affinity accelerates drug discovery by prioritizing compounds for experimental validation. Current methods face three limitations: sequence-based approaches discard spatial geometric constraints, structure-based methods fail to exploit hydrogen bond features, and conventional loss functions neglect prediction-target correlation, a key factor for identifying high-affinity compounds in virtual screening. We developed HBGSA (Hydrogen Bond Graph with Self-Attention), a 3.06M-parameter model that encodes hydrogen bond spatial features. HBGSA uses graph neural networks to model hydrogen bond spatial topology with self-attention enhancement and Pearson correlation loss. Experimental results on PDBbind Core Set and CSAR-HiQ dataset demonstrate that HBGSA outperforms baseline methods with strong generalization capability. Ablation studies confirm the effectiveness of hydrogen bond modeling and Pearson correlation loss.

Junxiao Kong, Chupei Tang, Di Wang +4

School of Mathematics and Statistics, Lanzhou University, 222 South Tianshui Road, 730000, Gansu, China · Department of Computer Science, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon Tong, Hong Kong 999077, China

Protein-Ligand Binding Affinity Prediction Graph Neural Networks Molecular Property Prediction Drug Discovery

Apr 21, 2026cs.LG

Graph-Theoretic Models for the Prediction of Molecular Measurements

Graph-theoretic approaches offer simplicity, interpretability, and low computational cost for molecular property prediction. Among these, the model proposed by Mukwembi and Nyabadza, based on the external activity

D(G)

and internal activity

ζ(G)

indices, achieved strong results on a small flavonoid dataset. However, its ability to generalize to larger and chemically diverse datasets has not been tested. This study evaluates the baseline

D(G)

-

ζ(G)

polynomial model on five benchmark datasets from MoleculeNet, covering biological activity (BACE, 1,513 molecules), lipophilicity (LogP synthetic, 14,610 molecules; LogP experimental, 753 molecules), aqueous solubility (ESOL, 1,128 molecules), and hydration free energy (SAMPL, 642 molecules). The baseline model achieves an average

R^2 = 0.24

, confirming limited transferability. To address this, a systematic enhancement framework is proposed, progressively incorporating Ridge regularization, additional graph descriptors, physicochemical properties, ensemble learning with Gradient Boosting, Lasso feature selection, and a hybrid approach combining topological indices with Morgan fingerprints. The enhanced models raise the average best

R^2

to 0.79, with individual improvements ranging from 165% to 274%. All improvements are statistically significant (

p < 0.001

). A direct comparison with a Graph Convolutional Network under identical experimental conditions shows that the enhanced classical models match or outperform deep learning on all five datasets. Comparison with the recent GNN+PGM hybrid of Djagba et al.\ further confirms competitiveness, with the enhanced models achieving the best results on two datasets and tying on one. The entire framework requires no GPU, trains in under five minutes, and uses only open-source tools, making it accessible for researchers in resource-limited settings.

Anna Niane, Prudence Djagba

African Institute for Mathematical Sciences (AIMS), Mbour, Senegal. · Lyman Briggs College, Michigan State University, East Lansing, Michigan, USA.

Molecular Property Prediction

Apr 17, 2026cs.LG

Evaluating the Progression of Large Language Model Capabilities for Small-Molecule Drug Design

Large Language Models (LLMs) have the potential to accelerate small molecule drug design due to their ability to reason about information from diverse sources and formats. However, their practical utility remains unclear due to the lack of benchmarks that reflect real-world scenarios. In this work, we introduce a suite of chemically-grounded tasks spanning molecular property prediction, molecular representation transformations, and molecular design. Importantly, we formulate these tasks as reinforcement learning (RL) environments, enabling a unified approach for evaluation and post-training. Across three model families, we find that frontier models are increasingly proficient at chemical tasks, but that there is significant room for improvement, especially in experimental settings with low data. Critically, we show that RL-based post-training can substantially improve performance. A smaller model post-trained on our environments becomes competitive with state-of-the-art frontier models, despite a significantly weaker base model. This suggests a practical route toward employing LLMs in drug discovery; by combining carefully-designed evaluation tasks with targeted post-training, we can both elucidate and close critical capability gaps.

Shriram Chennakesavalu, Kirill Shmilovich, Hayley Weir +5

Prescient Design, Genentech, South San Francisco, CA, USA

LLM Evaluation Molecular Property Prediction LLM Post-Training Drug Discovery