Papers
arxiv:2605.09100

GRC: Unifying Reasoning-Driven Generation, Retrieval and Compression

Published on May 12
Authors:
,
,

Abstract

A unified training framework integrates reasoning-driven generation, text representation, and context compression tasks in a single forward pass using meta latent tokens and specialized tuning approaches.

Text embedding and generative tasks are usually trained separately based on large language models (LLMs) nowadays. This causes a large amount of training cost and deployment effort. Context compression is also a challenging and pressing task, which is vital to reasoning-driven generation, and agentic tasks requiring long context and continual learning. In this paper, we explore how to unify reasoning-driven generation, reasoning-enhanced text representation and context compression tasks in one forward pass for LLMs. Through meta latent tokens and a unified generative, representative and compressive tuning approach, we propose a training framework named GRC that bridges the three tasks. The trained models can accomplish three objectives in a single forward pass while maintaining modular, LEGO-style flexibility during inference. This design greatly reduces the deployment effort for retrieval-augmented generation (RAG) and achieves efficient inference and three times data utilization during training. Furthermore, this framework design enables a new paradigm for text embedding: self-reason-latent embeds, and a new generation paradigm, latent memory-augmented generation, where compressed and internalized KV cache with O(1) length is used as the updatable memory. We also propose hybrid paged attention to speed up the inference of our models. Extensive experiments on reasoning-intensive retrieval benchmarks, generative tasks, document compression, latency evaluation, and RAG settings demonstrate the effectiveness of our method and may shed light on the truly unified model that can handle reasoning-driven generation, embedding and compression tasks seamlessly.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2605.09100
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2605.09100 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2605.09100 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2605.09100 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.