REVIEW 16 cited by
Nature Language Model: Deciphering the Language of Nature for Scientific Discovery
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
Nature Language Model: Deciphering the Language of Nature for Scientific Discovery
read the original abstract
Foundation models have revolutionized natural language processing and artificial intelligence, significantly enhancing how machines comprehend and generate human languages. Inspired by the success of these foundation models, researchers have developed foundation models for individual scientific domains, including small molecules, materials, proteins, DNA, RNA and even cells. However, these models are typically trained in isolation, lacking the ability to integrate across different scientific domains. Recognizing that entities within these domains can all be represented as sequences, which together form the "language of nature", we introduce Nature Language Model (NatureLM), a sequence-based science foundation model designed for scientific discovery. Pre-trained with data from multiple scientific domains, NatureLM offers a unified, versatile model that enables various applications including: (i) generating and optimizing small molecules, proteins, RNA, and materials using text instructions; (ii) cross-domain generation/design, such as protein-to-molecule and protein-to-RNA generation; and (iii) top performance across different domains, matching or surpassing state-of-the-art specialist models. NatureLM offers a promising generalist approach for various scientific tasks, including drug discovery (hit generation/optimization, ADMET optimization, synthesis), novel material design, and the development of therapeutic proteins or nucleotides. We have developed NatureLM models in different sizes (1 billion, 8 billion, and 46.7 billion parameters) and observed a clear improvement in performance as the model size increases.
Forward citations
Cited by 16 Pith papers
-
BioMatrix: Towards a Comprehensive Biological Foundation Model Spanning the Modality Matrix of Sequences, Structures, and Language
BioMatrix unifies sequences, structures, and language for molecules and proteins inside one decoder-only foundation model via shared discrete tokens and achieves SOTA or competitive results on 77 of 80 downstream tasks.
-
VibeProteinBench: An Evaluation Benchmark for Language-interfaced Vibe Protein Design
VibeProteinBench is a new benchmark evaluating LLMs on open-ended language-interfaced protein design across recognition, engineering, and generation, with no model showing strong performance in all areas.
-
VibeProteinBench: An Evaluation Benchmark for Language-interfaced Vibe Protein Design
VibeProteinBench is a three-stage language-interfaced benchmark revealing that no current LLM performs strongly across recognition, engineering, and generation of proteins.
-
RosettaSearch: Multi-Objective Inference-Time Search for Protein Sequence Design
RosettaSearch applies LLM-driven multi-objective search at inference time to improve backbone-conditioned protein sequences, recovering designs with 18-68% better structural fidelity and 2.5x higher success rates than...
-
AlphaEvolve: A coding agent for scientific and algorithmic discovery
AlphaEvolve is an LLM-orchestrated evolutionary coding agent that discovered a 4x4 complex matrix multiplication algorithm using 48 scalar multiplications, the first improvement over Strassen's algorithm in 56 years, ...
-
TheBioCollection: Unified Pre-Training Scale LLM Corpus for Biology
A 52.6B-token multi-domain biology pretraining corpus with tool enrichment and new binding/localization instructions doubles a fixed base LLM's matched biology-eval score with little language forgetting.
-
SciCodePile: A 128GB Corpus and Executable Benchmark for Challenging Scientific Code Generation
SciCodePile contributes a 128GB scientific code corpus and a 200-task executable benchmark showing that current LLMs still fall far short of reliable scientific code generation.
-
PertReason: A Knowledge-Grounded Benchmark and Framework for Cell-State-Conditioned Mechanistic Reasoning of Perturbation Effects
PertReasonQA scores AI models on cell-state-conditioned mechanistic reasoning about perturbation effects, and PertReasonLM, trained with reasoning supervision, reaches 0.736 balanced accuracy and 0.976 edge recall ver...
-
Monkey King Bang: A Unified Scientific Multimodal Foundation Model
A shared-backbone multimodal model integrating six scientific modalities with native-output decoders shows competitive understanding and generation across biology, chemistry, weather, and medical imaging.
-
TheBioCollection: Unified Pre-Training Scale LLM Corpus for Biology
TheBioCollection, a 52.6B-token unified biology corpus with tool-computed text and new instruction tasks, raises a fixed 16B LLM's score on its matched biology eval from 0.223 to 0.499 (2.24×).
-
Automated Data Readiness for Scientific AI
REDI automates raw-to-AI-ready scientific data prep via a five-stage IPTSO pipeline with provenance, readiness scoring, and multi-domain validation matching expert references.
-
AMix-2: Establishing Protein as a Native Modality in Large Language Models
AMix-2 unifies protein sequences and text in one LLM via shared tokens and block-wise diffusion modeling, introduces the ProteinArena benchmark, and reports competitive performance against task-specific protein models...
-
SciCustom: A Framework for Custom Evaluation of Scientific Capabilities in Large Language Models
SciCustom constructs application-specific benchmarks for LLM scientific capabilities from large-scale data using ontology-grounded units, automated tagging, consensus retrieval, and proxy selection.
-
LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning
LatentChem reasons in continuous latent space for chemistry, achieving a 59.88% non-tie win rate over explicit CoT on ChemCoTBench with a 10.84x average reduction in reasoning overhead.
-
When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs
A new plausibility metric and a 6.4M-reaction dataset are proposed for benchmarking LLM retrosynthesis, and a model trained on that dataset scores highest under the metric.
-
Discovery and recovery of crystalline materials with property-conditioned transformers
Conditioning the attention layers of a crystal-writing transformer on continuous property values enables XRD-based structure recovery and targeted generation of photovoltaic candidates.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.