paper-with-me

홈 › Papers

Predict and Reconstruct: Joint Objectives for Self-Supervised Language Representation Learning

2026-04-16 · Aimen Boukhari arxiv

Masked language modelling (MLM) has been the dominant pre-training objective for text encoders since BERT, yet it encourages representations that are strongly anchored to surface-form token identity rather than deeper semantic structure. Inspired by the success of Joint Embedding Predictive Architectures (JEPA) (LeCun, 2022) in vision and audio, we propose a hybrid pre-training objective that combines a JEPA-style latent-space prediction loss with a standard MLM objective over a single shared encoder. A learnable scalar parameter continuously balances the two objectives during training. We pre-train both a hybrid model and a pure-MLM baseline on English Wikipedia using identical architectures and compute budgets (NVIDIA H100). Extensive representation analysis across five GLUE benchmarks (SST-2, MRPC, MNLI, CoLA, STS-B) using four pooling strategies reveals that the hybrid encoder produces significantly more uniform embeddings (uniformity less than -0.16 vs -0.05 for MLM), exhibits richer spectral geometry under max pooling, encodes less surface-level lexical information, and achieves a better semantic-to-lexical balance. Despite similar linear-probe downstream accuracy, the geometric differences are consistent and significant, suggesting that the JEPA predictive objective reshapes the latent space in ways that standard accuracy metrics alone cannot capture.

📄 PDF Abstract BibTeX arXiv:2606.05173

Code (0)

등록된 구현이 없습니다.

Tasks

Representation LearningLanguage Modelling

Similar Papers 제목 키워드 기반

OLIVE: View-Augmented Latent Prediction with Waveform Reconstruction for Speech SSL

2026-06-29 · Karl El Hajal, Mathew Magimai. -Doss arxiv

We propose Online Latent prediction with Invariant Views and rEconstruction (OLIVE), a self-supervised speech representation learning framework that jointly optimizes analysis and synthesis objectives. OLIVE combines vie…

Representation Learning

Objectives Matter: Understanding the Impact of Self-Supervised Objectives on Vision Transformer Representations

2023-04-25 · Shashank Shekhar, Florian Bordes, Pascal Vincent, Ari Morcos

Joint-embedding based learning (e.g., SimCLR, MoCo, DINO) and reconstruction-based learning (e.g., BEiT, SimMIM, MAE) are the two leading paradigms for self-supervised learning of vision transformers, but they differ sub…

Self-Supervised LearningSpecificity

Masked and Predictive Self-Supervised Foundation Models for 3D Brain MRI

2026-06-11 · Esra Ergün, Hersh Chandarana, Dan Sodickson, Gözde Ünal arxiv

Self-supervised foundation models have shown strong promise in medical imaging. However, existing MRI foundation-model studies have primarily emphasized segmentation and dense prediction tasks, while systematic investiga…

Representation Learning

Semi-supervised Pathology Segmentation with Disentangled Representations

2020-09-05 · Haochuan Jiang, Agisilaos Chartsias, Xinheng Zhang, Giorgos Papanastasiou 외

Automated pathology segmentation remains a valuable diagnostic tool in clinical practice. However, collecting training data is challenging. Semi-supervised approaches by combining labelled and unlabelled data can offer a…

AnatomyDiagnosticDisentanglementSegmentation

MJEPA: A Simple and Scalable Joint-Embedding Predictive Architecture for Audio-Visual Learning

2026-06-23 · Revant Teotia, Adrien Bardes, Michael Rabbat, Sumit Chopra 외 arxiv

Self-supervised learning from large-scale video data has emerged as a dominant paradigm for visual representation learning. Since audio and visual streams naturally co-occur in video data, extending this success to joint…

Self-Supervised LearningRepresentation Learning