RxnNano:Training Compact LLMs for Chemical Reaction and Retrosynthesis Prediction via Hierarchical Curriculum Learning
- URL: http://arxiv.org/abs/2603.02215v1
- Date: Tue, 10 Feb 2026 07:42:23 GMT
- Title: RxnNano:Training Compact LLMs for Chemical Reaction and Retrosynthesis Prediction via Hierarchical Curriculum Learning
- Abstract summary: We argue that the core challenge lies in instilling these knowledge into the models.<n>We propose a unified framework that prioritizes chemical understanding over scale through three key innovations.
- Score: 20.904697361501174
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Chemical reaction prediction is pivotal for accelerating drug discovery and synthesis planning. Despite advances in data-driven models, current approaches are hindered by an overemphasis on parameter and dataset scaling. Some methods coupled with evaluation techniques that bypass fundamental challenges in reaction representation and fail to capture deep chemical intuition like reaction common sense and {topological atom mapping logic}. We argue that the core challenge lies in instilling these knowledge into the models. To this end, we propose a unified framework that prioritizes chemical understanding over scale through three key innovations: (1) a {Latent Chemical Consistency} objective that models reactions as movements on a continuous chemical manifold, ensuring reversible and physically plausible transformations; (2) a {Hierarchical Cognitive Curriculum} that trains the model through progressive stages, from syntax mastery to semantic reasoning, building robust chemical intuition; (3) {Atom-Map Permutation Invariance (AMPI)}, which force the model to learn invariant relational topology and balance multi-task learning. (4)and structured plan-based reasoning to improve the performance of the LLMs. Our compact {0.5B-parameter model}, \textbf{RxnNano} significantly outperforms fine-tuned LLMs ten times larger (>7B) and all the domain baselines, achieving a 23.5\% Top-1 accuracy improvement on rigorous benchmarks without test-time augmentation. https://github.com/rlisml/RxnNano.
Related papers
- Influence-Guided Symbolic Regression: Scientific Discovery via LLM-Driven Equation Search with Granular Feedback [56.69850045068714]
We introduce textitInfluence-Guided Regression (IGSR), a method that frames equation discovery as an iterative two-step process.<n>We demonstrate IGSRs effectiveness on a diverse suite of benchmarks, including LLM-SRBench, pharmacological PKPD models, an epidemiological simulation, and real-world genomic data.
arXiv Detail & Related papers (2026-05-27T23:48:01Z) - Molecular Identifier Visual Prompt and Verifiable Reinforcement Learning for Chemical Reaction Diagram Parsing [52.825281124618535]
Reaction diagram parsing (RxnDP) is critical for extracting chemical synthesis information from literature.<n>Recent Vision-Language Models (VLMs) have emerged as a promising paradigm to automate this complex visual reasoning task.<n>This work enhances VLM-based RxnDP from two complementary perspectives: prompting representation and learning paradigms.
arXiv Detail & Related papers (2026-03-16T09:17:05Z) - Modular Multi-Task Learning for Chemical Reaction Prediction [1.443416244644791]
Low-Rank Adaptation (LoRA) is a parameter-efficient alternative to full fine-tuning for organic reaction prediction.<n>LoRA achieves accuracy comparable to full fine-tuning while effectively mitigating catastrophic forgetting and better preserving multi-task performance.
arXiv Detail & Related papers (2026-02-11T01:17:06Z) - A Scientific Reasoning Model for Organic Synthesis Procedure Generation [12.609346156252393]
We present QFANG, a scientific reasoning language model capable of generating precise, structured experimental procedures.<n>We introduce a Chemistry-Guided Reasoning (CGR) framework that produces chain-of-thought data grounded in chemical knowledge at scale.<n>We apply Reinforcement Learning from Verifiable Rewards (RLVR) to further enhance procedural accuracy.
arXiv Detail & Related papers (2025-12-15T18:55:39Z) - oMeBench: Towards Robust Benchmarking of LLMs in Organic Mechanism Elucidation and Reasoning [44.36582860924775]
We introduce oMeBench, the first large-scale, expert-curated benchmark for organic mechanism reasoning in organic chemistry.<n>We also propose oMeS, a dynamic evaluation framework that combines step-level logic and chemical similarity.
arXiv Detail & Related papers (2025-10-09T03:13:31Z) - Round-trip Reinforcement Learning: Self-Consistent Training for Better Chemical LLMs [51.29260537017623]
Large Language Models (LLMs) are emerging as versatile foundation models for computational chemistry.<n>These models often lack round-trip consistency.<n>We introduce Round-Trip Reinforcement Learning (RTRL), a novel framework that trains a model to improve its consistency.
arXiv Detail & Related papers (2025-10-01T23:58:58Z) - Interpretable Deep Learning for Polar Mechanistic Reaction Prediction [43.95903801494905]
We introduce PMechRP (Polar Mechanistic Reaction Predictor), a system that trains machine learning models on the PMechDB dataset.<n>We train compare a range of machine learning models, including transformer-based, graph-based and two-step siamese architectures.<n>Our best-performing approach was a hybrid model, which combines a 5-ensemble of Chemformer models with a two-step Siamese framework.
arXiv Detail & Related papers (2025-04-22T02:31:23Z) - FINEREASON: Evaluating and Improving LLMs' Deliberate Reasoning through Reflective Puzzle Solving [90.88021670297664]
FINEREASON is a logic-puzzle benchmark for evaluation of large language models' reasoning capabilities.<n>We introduce two tasks: state checking, and state transition, for a comprehensive evaluation of how models assess the current situation and plan the next move.<n>We show that models trained on our state checking and transition data demonstrate gains in math reasoning by up to 5.1% on GSM8K.
arXiv Detail & Related papers (2025-02-27T16:23:25Z) - Chemist-aligned retrosynthesis by ensembling diverse inductive bias models [5.47805641978534]
RetroChimera is a frontier retrosynthesis model built upon two newly developed components with complementary inductive biases.<n>We show it outperforms all major models by a large margin, demonstrating robustness outside the training data.<n>We also demonstrate zero-shot transfer to an internal dataset from a major pharmaceutical company.
arXiv Detail & Related papers (2024-12-06T18:55:19Z) - Learning Chemical Reaction Representation with Reactant-Product Alignment [50.28123475356234]
RAlign is a novel chemical reaction representation learning model for various organic reaction-related tasks.<n>By integrating atomic correspondence between reactants and products, our model discerns the molecular transformations that occur during the reaction.<n>We introduce a reaction-center-aware attention mechanism that enables the model to concentrate on key functional groups.
arXiv Detail & Related papers (2024-11-26T17:41:44Z) - Retrosynthesis prediction enhanced by in-silico reaction data
augmentation [66.5643280109899]
We present RetroWISE, a framework that employs a base model inferred from real paired data to perform in-silico reaction generation and augmentation.
On three benchmark datasets, RetroWISE achieves the best overall performance against state-of-the-art models.
arXiv Detail & Related papers (2024-01-31T07:40:37Z) - Retrosynthesis Prediction with Conditional Graph Logic Network [118.70437805407728]
Computer-aided retrosynthesis is finding renewed interest from both chemistry and computer science communities.
We propose a new approach to this task using the Conditional Graph Logic Network, a conditional graphical model built upon graph neural networks.
arXiv Detail & Related papers (2020-01-06T05:36:57Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.