Pith. sign in

REVIEW 11 cited by

Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup

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

arxiv 2101.06983 v2 pith:CFWLZOAV submitted 2021-01-18 cs.LG cs.CLcs.IR

Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup

classification cs.LG cs.CLcs.IR
keywords batchcontrastivelearninglossmemorynegativesencoderexample
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Contrastive learning has been applied successfully to learn vector representations of text. Previous research demonstrated that learning high-quality representations benefits from batch-wise contrastive loss with a large number of negatives. In practice, the technique of in-batch negative is used, where for each example in a batch, other batch examples' positives will be taken as its negatives, avoiding encoding extra negatives. This, however, still conditions each example's loss on all batch examples and requires fitting the entire large batch into GPU memory. This paper introduces a gradient caching technique that decouples backpropagation between contrastive loss and the encoder, removing encoder backward pass data dependency along the batch dimension. As a result, gradients can be computed for one subset of the batch at a time, leading to almost constant memory usage.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Bekko Embedding: Parameter-Efficient Multilingual Retrieval with Ultra-Compact Encoders

    cs.IR 2026-07 conditional novelty 7.0

    Bekko a8m, with 7.7M active parameters, scores 56.2 on MMTEB Multilingual v2 Retrieval, beating mE5 models and BGE-M3, while a25m reaches 57.5, on par with gte-multilingual-base.

  2. Unified Work Embeddings: Contrastive Learning of a Bidirectional Multi-task Ranker

    cs.CL 2025-11 unverdicted novelty 7.0

    UWE is a task-agnostic bi-encoder that uses many-to-many InfoNCE and token-level soft late interaction to achieve zero-shot ranking across unseen work-related target spaces while using far fewer parameters than Qwen3-...

  3. VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks

    cs.CV 2024-10 conditional novelty 7.0

    VLM2Vec converts state-of-the-art vision-language models into universal multimodal embedders via contrastive training on the new MMEB benchmark, delivering 10-20% absolute gains over prior models on both in-distributi...

  4. C-Pack: Packed Resources For General Chinese Embeddings

    cs.CL 2023-09 accept novelty 7.0

    C-Pack releases a new Chinese embedding benchmark, large training dataset, and optimized models that outperform priors by up to 10% on C-MTEB while also delivering English SOTA results.

  5. Illuminating Visual Identity in Universal Multimodal Embeddings

    cs.CV 2026-08 conditional novelty 6.0

    By adding identity-aware sampling and a contrastive loss on a new 28-dataset benchmark, the authors build multimodal embeddings that are far better at visual identity matching without losing general retrieval accuracy.

  6. Florence: A New Foundation Model for Computer Vision

    cs.CV 2021-11 unverdicted novelty 6.0

    Florence is a new vision foundation model that learns universal visual-language representations from web-scale data and reports state-of-the-art results on 44 benchmarks including 83.74% zero-shot ImageNet top-1 accuracy.

  7. Benchmarking Patent Embeddings: A Multi-Task Evaluation of 22 Models Across Retrieval, Classification, and Clustering

    cs.IR 2026-05 unverdicted novelty 5.0

    Multi-task evaluation of 22 patent embedding models finds task-specific fine-tuning benefits and significant cross-landscape retrieval degradation that cannot be fixed by hybrid fusion.

  8. Legal Retrieval for Public Defenders

    cs.IR 2026-01 conditional novelty 5.0

    NJ BriefBank is a domain-adapted legal retrieval tool for public defenders that improves on standard benchmarks by incorporating legal reasoning, domain data, and synthetic examples, with a new released taxonomy and a...

  9. Improving Korean-English Cross-Lingual Retrieval: A Data-Centric Study of Language Composition and Model Merging

    cs.IR 2025-07 unverdicted novelty 5.0

    Language composition in training data creates opposing effects on CLIR and mono-IR performance for Korean-English retrieval, which model merging can partially resolve.

  10. VLM2Vec-V2: Advancing Multimodal Embedding for Videos, Images, and Visual Documents

    cs.CV 2025-07 unverdicted novelty 5.0

    VLM2Vec-V2 is a multimodal embedding model trained on an extended MMEB-V2 benchmark that adds video and visual document tasks and reports gains on both new and prior image benchmarks.

  11. Unified Supervision for Walmart's Sponsored Search Retrieval via Joint Semantic Relevance and Behavioral Engagement Modeling

    cs.IR 2026-04 unverdicted novelty 4.0

    A hybrid supervision method for bi-encoder retrievers combines graded relevance from teacher models, production retrieval priors, and selective engagement to improve relevance and NDCG over Walmart's current sponsored...