Related papers: ReinPool: Reinforcement Learning Pooling Multi-Vec…

Reinforcement learning with verifiable rewards (RLVR) has advanced the reasoning capabilities of large language models. However, existing methods rely solely on outcome rewards, without explicitly optimizing verification or leveraging…

In this paper, we propose an end-to-end Retrieval-Augmented Visual Language Model (REVEAL) that learns to encode world knowledge into a large-scale memory, and to retrieve from it to answer knowledge-intensive queries. REVEAL consists of…

We present ReMatch, a framework that leverages the generative strength of MLLMs for multimodal retrieval. Previous approaches treated an MLLM as a simple encoder, ignoring its generative nature, and under-utilising its compositional…

Word embeddings are commonly used as a starting point in many NLP models to achieve state-of-the-art performances. However, with a large vocabulary and many dimensions, these floating-point representations are expensive both in terms of…

Multimodal documents contain diverse elements, such as tables, figures, and layouts, which can complicate retrieval tasks. While current approaches typically combine dense visual embedding models with supervised rerankers to achieve…

Pooling is a critical operation in convolutional neural networks for increasing receptive fields and improving robustness to input variations. Most existing pooling operations downsample the feature maps, which is a lossy process. Moreover,…

Dense retrieval has achieved impressive advances in first-stage retrieval from a large-scale document collection, which is built on bi-encoder architecture to produce single vector representation of query and document. However, a document…

In continual learning, a model learns incrementally over time while minimizing interference between old and new tasks. One of the most widely used approaches in continual learning is referred to as replay. Replay methods support interleaved…

Human memory is inherently prone to forgetting. To address this, multimodal embedding models have been introduced, which transform diverse real-world data into a unified embedding space. These embeddings can be retrieved efficiently, aiding…

Learned multivector representations power modern search systems with strong retrieval effectiveness, but their real-world use is limited by the high cost of exhaustive token-level retrieval. Therefore, most systems adopt a…

Developing a single foundation model with the capability to excel across diverse tasks has been a long-standing objective in the field of artificial intelligence. As the wave of general-purpose foundation models sweeps across various…

We propose ReinFlow, a simple yet effective online reinforcement learning (RL) framework that fine-tunes a family of flow matching policies for continuous robotic control. Derived from rigorous RL theory, ReinFlow injects learnable noise…

We develop a portfolio allocation framework that leverages deep learning techniques to address challenges arising from high-dimensional, non-stationary, and low-signal-to-noise market information. Our approach includes a dynamic embedding…

Modern deep learning-based recommendation systems exploit hundreds to thousands of different categorical features, each with millions of different categories ranging from clicks to posts. To respect the natural diversity within the…

Embedding-based retrieval methods construct vector indices to search for document representations that are most similar to the query representations. They are widely used in document retrieval due to low latency and decent recall…

On-policy distillation is pivotal for transferring reasoning capabilities to capacity-constrained models, yet remains prone to instability and negative transfer. We show that on-policy distillation can be interpreted, both theoretically and…

Multimodal retrieval over text corpora remains a fundamental challenge: the best vision-language encoder achieves only 27.6 nDCG@10 on MM-BRIGHT, a reasoning-intensive multimodal retrieval benchmark, underperforming strong text-only…

In deep learning, embeddings are widely used to represent categorical entities such as words, apps, and movies. An embedding layer maps each entity to a unique vector, causing the layer's memory requirement to be proportional to the number…

Embedding words in a vector space has gained a lot of attention in recent years. While state-of-the-art methods provide efficient computation of word similarities via a low-dimensional matrix embedding, their motivation is often left…

Multi-vector retrieval methods, exemplified by the ColBERT architecture, have shown substantial promise for retrieval by providing strong trade-offs in terms of retrieval latency and effectiveness. However, they come at a high cost in terms…