REVIEW 13 cited by
Translation between Molecules and Natural Language
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
We present $\textbf{MolT5}$ $-$ a self-supervised learning framework for pretraining models on a vast amount of unlabeled natural language text and molecule strings. $\textbf{MolT5}$ allows for new, useful, and challenging analogs of traditional vision-language tasks, such as molecule captioning and text-based de novo molecule generation (altogether: translation between molecules and language), which we explore for the first time. Since $\textbf{MolT5}$ pretrains models on single-modal data, it helps overcome the chemistry domain shortcoming of data scarcity. Furthermore, we consider several metrics, including a new cross-modal embedding-based metric, to evaluate the tasks of molecule captioning and text-based molecule generation. Our results show that $\textbf{MolT5}$-based models are able to generate outputs, both molecules and captions, which in many cases are high quality.
Forward citations
Cited by 13 Pith papers
-
Atomistic Language Models Understand and Generate Materials
ALMs unify pretrained atomistic encoder, LLM, and denoising diffusion via continuous projectors and staged training to reach SOTA on text-conditioned crystal prediction and de novo generation.
-
MoleCode unlocks structural intelligence in large language models
MoleCode is a training-free, LLM-native representation that makes molecular graphs with explicit atoms, bonds, and topology directly readable and editable in language models, improving structural tasks over implicit s...
-
Back to Basics: Improving Molecular Understanding in LLMs via SMILES-Graph Translation
Training LLMs first on bidirectional SMILES–graph conversion plus progressive CoT yields large structure-perception gains and better property prediction and molecular optimization.
-
OPD-Evolver: Cultivating Holistic Agent Evolver via On-Policy Distillation
OPD-Evolver uses on-policy self-distillation in fast interaction and slow attribution loops to build agents with holistic memory competence, outperforming prior systems by up to 11.5% and allowing a 9B model to compet...
-
Mol-Debate: Multi-Agent Debate Improves Structural Reasoning in Molecular Design
Mol-Debate applies multi-agent debate in an iterative loop with perspective orchestration to achieve state-of-the-art text-guided molecular design, scoring 59.82% exact match on ChEBI-20 and 50.52% weighted success on...
-
IR-Agent: Expert-Inspired LLM Agents for Structure Elucidation from Infrared Spectra
IR-Agent is a multi-agent LLM framework that emulates expert IR spectral analysis procedures to improve molecular structure elucidation accuracy and adaptability.
-
SciCore-Mol: Augmenting Large Language Models with Pluggable Molecular Cognition Modules
SciCore-Mol augments LLMs with three integrated modules for molecular perception, latent diffusion generation, and reaction reasoning, claiming an 8B open model competes with or exceeds proprietary systems on chemical tasks.
-
NMIRacle: Multi-modal Generative Molecular Elucidation from IR and NMR Spectra
NMIRacle generates molecular structures from combined raw IR and NMR spectra, improving Top-1 elucidation accuracy from 0.41 to 0.48 over the NMR2Struct baseline on the Alberts benchmark.
-
An Encoder-Decoder Foundation Chemical Language Model for Generative Polymer Design
A T5-based polymer language model pre-trained on 100 million hypothetical polymers predicts thermal, electronic, and solubility properties and generates polymers conditioned on a target glass-transition temperature, w...
-
HSA-Net: Hierarchical and Structure-Aware Framework for Efficient and Scalable Molecular Language Modeling
HSA-Net improves molecular language modeling by adaptively switching between cross-attention and Mamba projectors across GNN layers and fusing the results with a sparse mixture-of-experts.
-
CROP: Integrating Topological and Spatial Structures via Cross-View Prefixes for Molecular LLMs
Cross-view prefix resampling, guided by the LLM's SMILES encoding, lets a Galactica-based model exploit molecular graphs and images at low context cost, improving captioning, IUPAC naming, and property prediction.
-
Adopting Reinforcement Learning with Verifiable Rewards for Molecular Generation
LLMol fine-tunes an LLM on simplified SELFIES and uses GRPO with RDKit-derived rewards for targeted molecular generation, but its own benchmark tables contradict the claimed state-of-the-art performance.
-
SmileyLlama: Modifying Large Language Models for Directed Chemical Space Exploration
SmileyLlama is an LLM transformed via SFT and DPO to generate valid novel drug-like molecules with user-specified properties and optimized 3D conformations for high binding affinity.
Discussion (0). Sign in to comment.