TriForces: Augmenting Atomistic GNNs for Transferable Representations

• URL: http://arxiv.org/abs/2605.20581v1
• Date: Wed, 20 May 2026 00:38:43 GMT
• Title: TriForces: Augmenting Atomistic GNNs for Transferable Representations
• Abstract summary: Machine learning interatomic potentials (MLIPs) achieve excellent accuracy when trained on large Density Functional Theory (DFT) data.<n>MLIPs transfer inconsistently across domains, with representations that often loose accessible composition and structure information.<n>We present TriForces, a model-agnostic three-stream framework that separates composition and structure information, combined with self-supervised learning to preserve transferable representations.
• Score: 41.22575447426028
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Machine learning interatomic potentials (MLIPs) achieve excellent accuracy when trained on large Density Functional Theory (DFT) data. To be useful in practice, they must often be adapted to target chemistries using small and expensive task-specific datasets. However, MLIPs transfer inconsistently across domains, with representations that often loose accessible composition and structure information. To address this, we present TriForces, a model-agnostic three-stream framework that separates composition and structure information, combined with self-supervised learning to preserve transferable representations. TriForces improves performance on MatBench and QM9 over baselines without needing DFT labels and enables efficient similar structure retrieval through its learned latent space. On OMat24, in limited-data training regime, TriForces reduces energy MAE by 57% at 20K samples only and improves force MAE across sample sizes. We release pretrained TriForces variants across multiple MLIP architectures with code at https://github.com/Ramlaoui/triforces.

Related papers

• DART-ing Through the Drift: Dynamic Tracing of Knowledge Neurons for Adaptive Inference-Time Pruning [6.3691159627915015]
  We introduce DART, a lightweight training-free method that performs on-the-fly context-based pruning.<n>DART monitors shifts in distributions to infer context changes, dynamically updating neuron-level masks to retain salient parameters.<n>It achieves accuracy gains of up to 14.5% on LLAMA-3.1-8B at 70% FFN sparsity and 3x better ROUGE-L scores with respect to static-masked pruning.
  arXiv  Detail & Related papers  (2026-01-30T06:48:16Z)
• LeMat-Traj: A Scalable and Unified Dataset of Materials Trajectories for Atomistic Modeling [34.31458248589154]
  We introduce LeMat-Traj, a curated dataset comprising over 120 million atomic configurations aggregated from large-scale repositories.<n>LeMat-Traj standardizes data representation, harmonizes results and filters for high-quality configurations across widely used DFT functionals.<n>We also present LeMaterial-Fetcher, a modular and open-source library designed to provide a reproducible framework for the community to easily incorporate new data sources.
  arXiv  Detail & Related papers  (2025-08-28T15:06:32Z)
• Mastering Collaborative Multi-modal Data Selection: A Focus on Informativeness, Uniqueness, and Representativeness [63.484378941471114]
  We propose a collaborative framework, DataTailor, which leverages three key principles--informativeness, uniqueness, and representativeness--for effective data selection.<n>Experiments on various benchmarks demonstrate that DataTailor achieves 101.3% of the performance of full-data fine-tuning with only 15% of the data.
  arXiv  Detail & Related papers  (2024-12-09T08:36:10Z)
• MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale [66.73529246309033]
  multimodal large language models (MLLMs) have shown significant potential in a broad range of multimodal tasks.<n>Existing instruction-tuning datasets only provide phrase-level answers without any intermediate rationales.<n>We introduce a scalable and cost-effective method to construct a large-scale multimodal instruction-tuning dataset with rich intermediate rationales.
  arXiv  Detail & Related papers  (2024-12-06T18:14:24Z)
• Building on Efficient Foundations: Effectively Training LLMs with Structured Feedforward Layers [16.253898272659242]
  State-of-the-art results in large language models (LLMs) often rely on scale, which becomes computationally expensive. Our study focuses on transformer-based LLMs, specifically targeting the computationally intensive feedforward networks (FFNs) We show that wide and structured networks can utilize training FLOPs more efficiently, with fewer parameters and lower loss than dense models at their optimal trade-off.
  arXiv  Detail & Related papers  (2024-06-24T08:43:21Z)
• Compute Better Spent: Replacing Dense Layers with Structured Matrices [77.61728033234233]
  We identify more efficient alternatives to dense matrices, as exemplified by the success of convolutional networks in the image domain. We show that different structures often require drastically different initialization scales and learning rates, which are crucial to performance. We propose a novel matrix family containing Monarch matrices, the Block-Train, which we show performs better than dense for the same compute on multiple tasks.
  arXiv  Detail & Related papers  (2024-06-10T13:25:43Z)
• FedLLM-Bench: Realistic Benchmarks for Federated Learning of Large Language Models [48.484485609995986]
  Federated learning has enabled multiple parties to collaboratively train large language models without directly sharing their data (FedLLM) There are currently no realistic datasets and benchmarks for FedLLM. We propose FedLLM-Bench, which involves 8 training methods, 4 training datasets, and 6 evaluation metrics.
  arXiv  Detail & Related papers  (2024-06-07T11:19:30Z)
• Towards Compact 3D Representations via Point Feature Enhancement Masked Autoencoders [52.66195794216989]
  We propose Point Feature Enhancement Masked Autoencoders (Point-FEMAE) to learn compact 3D representations. Point-FEMAE consists of a global branch and a local branch to capture latent semantic features. Our method significantly improves the pre-training efficiency compared to cross-modal alternatives.
  arXiv  Detail & Related papers  (2023-12-17T14:17:05Z)
• Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training [44.790636524264]
  Point Prompt Training is a novel framework for multi-dataset synergistic learning in the context of 3D representation learning. It can overcome the negative transfer associated with synergistic learning and produce generalizable representations. It achieves state-of-the-art performance on each dataset using a single weight-shared model with supervised multi-dataset training.
  arXiv  Detail & Related papers  (2023-08-18T17:59:57Z)
• ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding [96.95120198412395]
  We introduce tri-modal pre-training framework that automatically generates holistic language descriptions for 3D shapes. It only needs 3D data as input, eliminating the need for any manual 3D annotations, and is therefore scalable to large datasets. We conduct experiments on two large-scale 3D datasets, NN and ShapeNet, and augment them with tri-modal datasets of 3D point clouds, captioning, and language for training. Experiments show that NN-2 demonstrates substantial benefits in three downstream tasks: zero-shot 3D classification, standard 3D classification with finetuning, and 3D (3D
  arXiv  Detail & Related papers  (2023-05-14T23:14:09Z)
• MASTER: Multi-task Pre-trained Bottlenecked Masked Autoencoders are Better Dense Retrievers [140.0479479231558]
  In this work, we aim to unify a variety of pre-training tasks into a multi-task pre-trained model, namely MASTER. MASTER utilizes a shared-encoder multi-decoder architecture that can construct a representation bottleneck to compress the abundant semantic information across tasks into dense vectors.
  arXiv  Detail & Related papers  (2022-12-15T13:57:07Z)

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.