The linear memory growth of the KV cache poses a significant bottleneck for LLM inference in long-context tasks. Existing static compression methods often fail to preserve globally important information. Although recent dynamic retrieval approaches attempt to address this issue, they typically suffer from coarse-grained caching strategies and incur high I/O overhead. To overcome these limitations, we propose HeteroCache, a training-free dynamic compression framework. Our method is built on two key insights: attention heads exhibit diverse temporal heterogeneity, and there is significant spatial redundancy among heads within the same layer.Guided by these insights, HeteroCache categorizes heads based on stability and similarity, applying a fine-grained weighting strategy that allocates larger cache budgets to heads with rapidly shifting attention to capture context changes.Furthermore, it features a hierarchical storage mechanism where representative heads monitor attention drift to trigger asynchronous, on-demand context retrieval, thereby hiding I/O latency.Experiments demonstrate that HeteroCache achieves state-of-the-art performance on long-context benchmarks and accelerates decoding by up to 3× compared to the original model with a 224K context. Our code is available at https://github.com/ponytaill/HeteroCache.

Procedural memory enables large language model (LLM) agents to internalize ”how-to” knowledge and thus reduce redundant trial-and-error. However, existing frameworks predominantly suffer from a ”passive accumulation” paradigm, treating memory as a static append-only archive. To bridge the gap between static storage and dynamic reasoning, we propose ReMe (Remember Me, Refine Me), a comprehensive framework for experience-driven agent evolution. ReMe manages the memory lifecycle via three mechanisms: 1) multi-faceted distillation, which extracts fine-grained experiences by recognizing success patterns, analyzing failure triggers and generating comparative insights; 2) context-adaptive reuse, which tailors historical insights to new contexts through scenario-aware indexing; and 3) utility-based refinement, which automatically adds validated memories and prunes outdated ones to maintain a compact, high-quality experience pool. Experiments on BFCL-V3 and AppWorld demonstrate that ReMe establishes a new state-of-the-art in agent memory system. Crucially, we observe a significant memory-scaling effect: Qwen3-8B equipped with ReMe outperforms larger, memoryless Qwen3-14B, indicating that self-evolving memory provides a computation-efficient path for lifelong learning.

This paper presents MaRSI, an automatic subject indexing method designed to address the limitations of traditional manual indexing and emerging GenAI technologies. Focusing on improving indexing accuracy in cross-lingual contexts and balancing efficiency and accuracy in large-scale datasets, MaRSI mimics human reference learning behavior by constructing semantic indexes from pre-indexed document. It calculates similarity to retrieve relevant references, merges, and reorders their topics to generate index results. Experiments demonstrate that MaRSI outperforms supervised fine-tuning of LLMs on the same dataset, offering advantages in speed, effectiveness, and interpretability.