KV Cache Translation across Heterogeneous Large Language Models
cs.CL
Submitted: 2026-07-31
Updated: 2026-10-03
Code: https://github.com/XiaoDu-flying/Interlat4https:
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- TurboRAG: Accelerating Retrieval-Augmented Generation with Precomputed KV Caches for Chunked Text
- Cache-to-Cache: Direct Semantic Communication Between Large Language Models
- KVComm: Enabling Efficient LLM Communication through Selective KV Sharing
- Enabling Agents to Communicate Entirely in Latent Space
- Latent Space Communication via K-V Cache Alignment
- Spurious Forgetting in Continual Learning of Language Models
- Deep Transformers without Shortcuts: Modifying Self-attention for Faithful Signal Propagation
- On the Convergence of Gradient Descent on Learning Transformers with Residual Connections
- Communication-Efficient On-Device Machine Learning: Federated Distillation and Augmentation under Non-IID Private Data
- Emergent Multi-Agent Communication in the Deep Learning Era
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- Vision Wormhole: Latent-Space Communication in Heterogeneous Multi-Agent Systems
- KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering