This paper proposes Supervised Memory Training (SMT), a method for training nonlinear RNNs that sidesteps recurrent credit propagation entirely.

The paper introduces COMET, a novel PLS-SVD framework, to analyze the audio-text modality gap in CLAP models, showing that shared concepts are captured by a small subset of axes, and proposes a spectr…

This paper presents the benefits of visual pretraining for foundation model intelligence, outperforming text-only pretraining on multiple backbones and benchmarks.

The paper proposes WEINCE, a modified InfoNCE objective that uses extreme value theory corrections to improve contrastive learning by more accurately modeling the selection of hard negative examples.

The paper introduces CERA, a novel contrastive retrieval framework that improves RAG factuality and interpretability by using subjectivity-based hard negative selection and an auxiliary attention alig…

BayesNCL introduces a probabilistic gating mechanism to resolve the optimization conflict in Contrastive Learning, leading to highly disentangled and semantically consistent representations.

The paper introduces Contrastive Reflection (CORE), a novel non-parametric method that rapidly improves language model reasoning by distilling contrasts between successful and unsuccessful problem att…

The paper proposes a multi-modal contrastive learning framework to improve the generalization of machine learning models in cybersecurity by transferring knowledge from rich textual vulnerability desc…

This paper explains how discarded norms in contrastive embedding models correlate with semantic properties and provides a theoretical framework.

The paper introduces the Temporal Contrastive Transformer (TCT) for financial crime detection, demonstrating that its self-supervised embeddings capture meaningful temporal behavioral patterns, though…

The paper proposes pretraining a Perceiver-style in-context learner on synthetic data to solve Multiple Instance Learning (MIL) tasks efficiently in the low-label regime.

The paper introduces BabyCL, a continual multimodal learning framework that processes egocentric video data in a single chronological pass, demonstrating that meaningful word-referent mappings can be…

The paper proposes In-Context Visual Contrastive Optimization (IC-VCO) to rigorously mitigate multimodal hallucinations in Vision-Language Models by optimizing contrastive learning within a shared mul…

PromptEmbedder introduces a dual-LLM framework that efficiently and transferably adapts text embeddings by decoupling task-specific knowledge from the backbone model, significantly reducing computatio…

The paper introduces RAG-Pref, a novel, training-free Retrieval Augmented Generation (RAG) method for preference alignment that significantly improves LLM refusal guardrails against agentic attacks wi…

This paper proposes Joint Speech-Text Interleaved Pretraining (JSTIP) for speech recognition, which constructs interleaved speech-text sequences and achieves consistent entity accuracy improvement.

This paper demonstrates that in-domain pretraining of BERT significantly improves the detection of DNS exfiltration, particularly in maintaining a low false positive rate.

The paper introduces and evaluates five parameter alignment strategies that significantly mitigate catastrophic forgetting when continually pretraining multilingual expert language models across multi…

This paper identifies modality-order sensitivity as a failure in vision-language models and introduces a test-time training method to mitigate it, resulting in improved performance.

The paper proposes AlignG, a method that learns context-conditioned predicate semantics by using prototype feedback to adapt relation representations based on image-specific evidence, significantly im…