arXiv Computation and Language September 23, 2026 By Simon P. Villani

LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay

Read the original on arXiv Computation and Language →

The paper introduces LatentPort, a method that allows a language model to transfer its live memory to another model without requiring the receiver to reread the context. Experiments on a Qwen3.5 4B-to-9B sibling pair show that adding a Gated DeltaNet (GDN) persistent-state package reduces negative log‑likelihood by 0.747 nats/token and improves performance across 64 PG19 documents. The study also demonstrates that direct recurrent and convolution reuse outperforms learned GDN maps, and a 434,176‑parameter correction further narrows the performance gap to the native 9B model.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI

Sep 1

Tail-Replay: Escaping the Curse of Linear Attention in Prefix Caching for Hybrid LLMs

arXiv:2608.30310v1 Announce Type: cross Abstract: Hybrid large language models interleave full-attention layers with linear-attention layers to reduce the cost of long-context inference. This structu...

By Yirui Liu, Ruoling Qi, Xuaner Wu, Penghang Liu, Jian Chen

llms benchmarks

arXiv Machine Learning

Aug 14

Subtract, Transport, or Replay? Auditable Deletion from Language-Model Memory

arXiv:2607. 27539v2 Announce Type: replace Abstract: Exact deletion from persistent language-model memory depends on whether a record's effect remains addressable after later computation.

By Vishwajith Ramesh

efficiency

arXiv AI

Sep 1

DASC: Decay-Aware State Compression for Hybrid Linear-Attention Serving

arXiv:2608.30386v1 Announce Type: cross Abstract: Hybrid linear-attention architectures have recently scaled to large open-weight models, offering quality competitive with full attention while substa...

By Yanqi Yu, Pingwei Sun, Jianchao Tan, Tao Zhang, Yuchen Xie, Xunliang Cai, Yao Liu

efficiency benchmarks

arXiv Machine Learning

Sep 22

GDN Tree-Scan: Served Tree Verification for Recurrent-Hybrid Language Models

arXiv:2609.23900v1 Announce Type: new Abstract: Tree speculative decoding verifies multiple candidate continuations in one target forward pass. For attention-only transformers, the verifier mainly ne...

By Zhiyuan Ma

llms efficiency safety

arXiv Computation and Language

Sep 7

What Attention Recalls and Recurrence Controls in Hybrid Language Models

Hybrid language models combine attention with a fixed-size recurrent state, yet the distinct roles of each component are not well understood. The authors introduce two cache-level interventions—split-prefill and state-swap—to isolate the contributions of the KV cache (attention) and the recurrent state. Experiments on Qwen3.5 and Falcon-H1 show that exact retrieval depends almost entirely on attention, while output language and persona rely mainly on recurrence, with the state-swap intervention confirming that answers derive from the KV side and language from the recurrent side.

By Kirill Afendulev, Alexey Dontsov, Elena Tutubalina, Anton Korznikov

llms efficiency

arXiv AI

Jul 24

Naju: A Native Discrete State-Space Model with Independent Retention and Writing for Long-Sequence Memory

arXiv:2607. 21000v1 Announce Type: new Abstract: Long-sequence memory tracking places two opposing demands on a recurrent state: near-lossless retention of stored bindings over long horizons, and active overwriting of stale ones.

By Hyuk Lim, Seunghyun Yoon

llms

One email a morning, machine-written

Related stories

Tail-Replay: Escaping the Curse of Linear Attention in Prefix Caching for Hybrid LLMs

Subtract, Transport, or Replay? Auditable Deletion from Language-Model Memory

DASC: Decay-Aware State Compression for Hybrid Linear-Attention Serving

GDN Tree-Scan: Served Tree Verification for Recurrent-Hybrid Language Models

What Attention Recalls and Recurrence Controls in Hybrid Language Models

Naju: A Native Discrete State-Space Model with Independent Retention and Writing for Long-Sequence Memory