Smiles
PulseAugur coverage of Smiles — every cluster mentioning Smiles across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New dataset OPRD-Imp aims to improve AI prediction of pharmaceutical synthesis impurities
Researchers have introduced OPRD-Imp, a new dataset designed to address the scarcity of data for predicting reaction impurities in pharmaceutical synthesis. This dataset comprises 229 organic reactions from 62 publicati…
-
New MECo Framework Translates LLM Intentions to Executable Code for Molecular Editing
Researchers have developed MECo, a new framework designed to improve molecular editing for drug discovery. MECo translates high-level editing intentions generated by large language models into executable code, ensuring …
-
LLMs fail to bind molecular knowledge across different identifiers, new study finds
A new research paper published on arXiv details a phenomenon called "cross-identifier binding failure" in large language models (LLMs) used for chemistry. The study found that even specialized chemistry LLMs struggle to…
-
New AI framework PairIT optimizes molecules for drug discovery
Researchers have developed PairIT, a new framework for autoregressive transformers designed to optimize molecules for drug discovery. This method leverages pairs of similar molecules to learn meaningful edits that impro…
-
New molecular language STAMP enables fragment-level learning and editing
Researchers have developed STAMP, a novel molecular language designed for fragment-level learning and editing in chemistry. Unlike existing methods, STAMP explicitly encodes motif identity, attachment sites, and tree to…
-
New framework enables controlled molecule generation for drug discovery
Researchers have developed a new framework called Cross-Modality Controlled Molecule Generation with Diffusion Language Model (CMCM-DLM) to generate molecules with specific properties. This modular approach extends pre-…
-
ChemMLLM: New multimodal LLM for chemical understanding and generation
Researchers have introduced ChemMLLM, a novel multimodal large language model designed for chemical understanding and generation. This model addresses the gap in cross-modal capabilities for chemical MLLMs by integratin…
-
Apple ML Research explores LLM agents, translation, speech, and safety
Apple's Machine Learning Research division has published several papers detailing advancements in AI. One paper introduces SelfCompact, a system that allows language model agents to autonomously manage their context by …
-
GraphNOSE: New Graph Transformer Predicts Olfactory Qualities
Researchers have developed GraphNOSE, an open-source graph transformer framework designed to predict olfactory qualities from molecular structures. This new model demonstrates superior performance compared to existing l…
-
Frontier LLMs show widespread verbatim retrieval of molecular data, study finds
A new study published on arXiv investigates the phenomenon of "molecular déjà vu" in frontier language models, where models appear to retrieve published molecular property values verbatim rather than predicting them. Th…
-
New method uses probabilistic model checking for AI sequence models
Researchers have developed a new pipeline that uses probabilistic model checking to analyze autoregressive neural sequence models, addressing limitations of traditional test-set accuracy. This method quantifies the prob…
-
AI uses language to guide crystal and molecular structure generation
Two new research papers introduce novel methods for structure generation using language-informed flow matching. The first, TFMat, uses structured text to guide crystal structure generation, improving match rates and ali…
-
New agentic LLM S3C-LLM enhances molecular structure elucidation
Researchers have developed S3C-LLM, a novel agentic language model designed for spectrum-to-structure elucidation in molecular analysis. Unlike previous methods that directly convert spectra to SMILES, S3C-LLM mimics th…
-
LLM-powered framework enhances molecular inverse design
Researchers have developed a closed-loop framework called \"method\" for molecular inverse design, which uses a large language model (LLM) to reason over task instructions, optimization history, and oracle feedback. Thi…
-
VERDICT system enhances chemical structure recognition accuracy
Researchers have developed a new system called VERDICT that improves the accuracy of Optical Chemical Structure Recognition (OCSR). VERDICT utilizes agreement among multiple recognizers, outperforming pixel-space verifi…
-
New VLSR Framework Enhances LLM Molecular Reasoning with Localization
Researchers have developed Visual Latent Structural Reasoning (VLSR), a new framework designed to improve how large language models (LLMs) understand molecular structures and predict their properties. Unlike previous me…
-
New molecular LLM uses substructure rationale for improved property prediction
Researchers have developed MR-MoL, a novel molecular large language model (LLM) designed for property prediction in drug discovery. Unlike existing models that represent molecules implicitly, MR-MoL explicitly incorpora…
-
MolBioKG system grounds unregistered molecules in biomedical knowledge graphs
Researchers have developed MolBioKG, a novel two-layer system designed to connect unregistered molecules to biomedical knowledge graphs. This system addresses the challenge of 'out-of-graph molecules' by using multi-res…
-
MinerU.Chem system achieves 93% accuracy in chemical structure recognition
A new system called MinerU.Chem has been developed to extract chemical structures and reactions from organic chemistry documents, converting them into machine-readable data. This system, integrated into the MinerU platf…
-
New model CheMatE unifies chemical structures and natural language
Researchers have developed CheMatE, a new embedding model designed to jointly represent chemical structures (SMILES) and natural language within a unified space. Built on a ModernBERT backbone, CheMatE employs a two-sta…