Related papers: MELODI: Exploring Memory Compression for Long Cont…

The quadratic complexity and indefinitely growing key-value (KV) cache of standard Transformers pose a major barrier to long-context processing. To overcome this, we introduce the Collaborative Memory Transformer (CoMeT), a novel…

There is a growing interest in expanding the input capacity of language models (LMs) across various domains. However, simply increasing the context window does not guarantee robust performance across diverse long-input processing tasks,…

Large language models (LLMs) achieve strong performance across a wide range of tasks, but remain frozen after pretraining until subsequent updates. Many real-world applications require timely, domain-specific information, motivating the…

Context-aware Machine Translation aims to improve translations of sentences by incorporating surrounding sentences as context. Towards this task, two main architectures have been applied, namely single-encoder (based on concatenation) and…

Long-horizon agentic reasoning necessitates effectively compressing growing interaction histories into a limited context window. Most existing memory systems serialize history as text, where token-level cost is uniform and scales linearly…

Sentence compression is the task of creating a shorter version of an input sentence while keeping important information. In this paper, we extend the task of compression by deletion with the use of contextual embeddings. Different from…

To enable embodied agents to operate effectively over extended timeframes, it is crucial to develop models that form and access memories to stay contextualized in their environment. In the current paradigm of training transformer-based…

Effective token compression remains a critical challenge for scaling models to handle increasingly complex and diverse datasets. A novel mechanism based on contextual reinforcement is introduced, dynamically adjusting token importance…

To support long-term interaction in complex environments, LLM agents require memory systems that manage historical experiences. Existing approaches either retain full interaction histories via passive context extension, leading to…

Large Language Models (LLMs) face significant challenges in long-context processing, including quadratic computational costs, information forgetting, and the context fragmentation inherent in retrieval-augmented generation (RAG). We propose…

This paper addresses the limitations of large language models in understanding long-term context. It proposes a model architecture equipped with a long-term memory mechanism to improve the retention and retrieval of semantic information…

The rise of AI-native Low-Code/No-Code (LCNC) platforms enables autonomous agents capable of executing complex, long-duration business processes. However, a fundamental challenge remains: memory management. As agents operate over extended…

Many learning tasks involve multi-modal data streams, where continuous data from different modes convey a comprehensive description about objects. A major challenge in this context is how to efficiently interpret multi-modal information in…

As large language models increasingly gain popularity in real-world applications, processing extremely long contexts, often exceeding the model's pre-trained context limits, has emerged as a critical challenge. While existing approaches to…

Repository-level code intelligence tasks require large language models (LLMs) to process long, multi-file contexts. Such inputs introduce three challenges: crucial context can be obscured by noise, truncated due to limited windows, and…

Recent advancements in large language models have significantly improved their context windows, yet challenges in effective long-term memory management remain. We introduce MemTree, an algorithm that leverages a dynamic, tree-structured…

In-memory columnar databases have become mainstream over the last decade and have vastly improved the fast processing of large volumes of data through multi-core parallelism and in-memory compression thereby eliminating the usual…

Long-context efficiency has recently become a trending topic in serving large language models (LLMs). And mixture of depths (MoD) is proposed as a perfect fit to bring down both latency and memory. In this paper, however, we discover that…

Long-horizon dialogue systems suffer from semanticdrift and unstable memory retention across extended sessions. This paper presents a Multi-Layer Memory Framework that decomposes dialogue history into working, episodic, and semantic layers…

Efficient long-context LLM deployment is stalled by a dichotomy between amortized compression, which struggles with out-of-distribution generalization, and Test-Time Training, which incurs prohibitive synthetic data costs and requires…