Skip to main content
Search

# Foundation Models

8 related articles

Subscribe to updates
  1. Hacker News

    Subvocal's glasses read silent mouthing with radio waves, no electrodes on your face

    Subvocal has moved silent speech input from face-taped electrodes to radio sensing: its glasses emit Bluetooth-like radio waves through the face, and silent mouthing of the tongue, lips and soft palate modulates them before foundation models decode the perturbations into text. The team says the electrode route suffered from sweat, electrical interference and placement variability, and that cross-user performance improved mainly by scaling pretraining data, with a demo user needing only a short 20-sentence calibration.

  2. arXiv

    A foundation model as critic backbone speeds multi-agent RL convergence for random access networks by at least 55%

    This work proposes a foundation-model-aided, fully decentralized multi-agent reinforcement learning framework in which a self-supervised forward-dynamics foundation model serves as a reward-agnostic critic backbone and devices exchange only scalar rewards via local consensus; it provides a finite-time convergence analysis and reports at least about 55% faster convergence than end-to-end training across fair-AoI, max-sum-rate, and fair-rate random access optimization tasks.

  3. arXiv

    BLS refines task vectors at test time, letting zero-shot policies surpass OLS and online adaptation baselines on OGBench, DMC, and HumEnv

    The authors propose BLS, an offline zero-shot task-inference method for Behavioral Foundation Models that starts from an ordinary-least-squares (OLS) task vector and refines it with a soft-margin contrastive loss plus a trust-region loss to reduce both reward reconstruction error and successor-measure mismatch; they prove a suboptimality gap upper bound characterized by successor-measure and reward residuals, and empirically BLS outperforms OLS, ZOL, ReLA, and LoLA on average across five feature representations on OGBench (proprioceptive and pixel observations), 20 DMC tasks, and 45 HumEnv humanoid control tasks, with negligible inference overhead.

  4. arXiv

    FairRSFM evaluates three remote sensing foundation models by biome group and finds aggregate metrics mask ecological performance gaps

    The authors introduce FairRSFM, which maps georeferenced samples from four remote sensing downstream datasets (m-EuroSAT, m-BigEarthNet, m-SA-Crop-Type, and MMEarth20K) to 14 terrestrial biome classes consolidated into six macro-groups, and evaluates Prithvi-EO-2.0, SatMAE, and DOFA under a unified frozen-backbone protocol across three random seeds, showing that aggregate metrics consistently mask biome-dependent disparities, for example Prithvi-EO-2.0 reaching high overall macro-F1 on m-EuroSAT while its worst-group score is markedly lower, and compares BOLP, DBR, and GroupDRO as mitigation baselines whose effectiveness is model- and task-dependent.

  5. OpenBox aligns 2D vision-foundation instance cues with 3D point clouds to produce more accurate and cheaper 3D box annotations without self-training across three autonomous-driving datasets

    OpenBox introduces a two-stage automatic annotation pipeline: in the first stage it uses a 2D vision foundation model and cross-modal instance alignment to associate instance-level cues from 2D images with the corresponding 3D point clouds; in the second stage it categorizes instances by rigidity and motion state and generates adaptive bounding boxes with class-specific size statistics, thereby producing high-quality 3D bounding box annotations without self-training and improving accuracy and efficiency over baselines on the Waymo Open Dataset, the Lyft Level 5 Perception dataset, and the nuScenes dataset.

  6. Researchers propose a research roadmap for merging search-based software engineering with AI foundation models across three directions

    Presented as a research roadmap, the work surveys the current relationship between search-based software engineering (SBSE), a field active for about 25 years, and AI foundation models (FMs) such as large language models, analyzing three core aspects—using FMs to enhance SBSE, applying SBSE to advance FMs, and exploring their integration—while identifying open challenges and potential research directions and envisioning the future of SBSE in the era of FMs.

  7. Deep Learning Foundation Models for Low-Data Regimes from Classical Molecular Descriptors

    This study introduces a new avenue for foundation model pretraining: supervised pretraining on low-noise, calculable molecular descriptors to obtain rich, highly transferable molecular representations, demonstrated with CheMeleon, an O(10M) parameter foundation model; across 58 benchmark data sets spanning properties relevant to small-molecule drug discovery and sourced from the industry-led Polaris benchmarking initiative, CheMeleon enables directed message-passing neural networks to finally exceed classical methods in the low-data regime, outperforms classical baselines such as Random Forest on molecular fingerprints and descriptors as well as existing foundation models under rigorous statistical comparisons, and the model and pretraining framework are open-sourced.

  8. Toward AI Virtual Cells for Hepatology: Representation, Generation, Dynamics, and Intervention in Single-Cell Models

    This review organizes current work toward an AI Virtual Cell (AIVC) for the liver into three complementary modeling routes—generative models that represent cell states, dynamics and transport models that infer state transitions, and pretrained or foundation models that test whether learned representations transfer across donors, etiologies, disease stages, and platforms—with perturbation-response prediction as a cross-cutting assessment, concluding that published models demonstrate only individual components such as atlas integration, inferred trajectories, transferable representations, and retrospective response programs, and do not yet constitute a prospectively validated liver simulator, so near-term use should prioritize experiment selection and hypothesis generation while clinical dec

You’re all caught up