Think Before You Speak: Enabling Training-Free Latent Communication Between Language Models
We introduce CacheFuse, a training-free method which demonstrates that augmenting a downstream model with an upstream model's KV cache using linear interpolation consistently enables superior performance than the standalone downstream model can achieve.
ProofPaper ↗
Key points
- Large language models (LLMs) have demonstrated remarkable capabilities in solving complex tasks in coding, mathematical logic, and iterative reasoning, especially by coordinating effectively within multi-agent systems (MAS).
- Recent work proposes that directly sharing the key-value (KV) cache between language models can enable efficient latent collaboration, entirely bypassing standard text-based communication.
- To enable latent communication between heterogeneous models of different sizes, prior work introduces trained fusion adapters for each model pair, which limits practical deployment.
- We systematically analyze CacheFuse across four reasoning and knowledge benchmarks and demonstrate up to 8.6 percentage points higher accuracy compared to single-model baselines and up to 12.2x TTFT speedup compared to text-based communication in multi-model inference.
Sources (1)
- [1]Think Before You Speak: Enabling Training-Free Latent Communication Between Language ModelsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 12, 04:40 AM
We introduce CacheFuse, a training-free method which demonstrates that augmenting a downstream model with an upstream model's KV cache using linear interpolation consistently enables superior performance than the standalone downstream model can achieve.
Large language models (LLMs) have demonstrated remarkable capabilities in solving complex tasks in coding, mathematical logic, and iterative reasoning, especially by coordinating effectively within multi-agent systems (MAS).
Extractive summary: sentences quoted from the sources.
Before this
- Oct 12, 2026AI Representations of Mental Health Stigma: Quantifying the Gap Between AI Estimations and Human Perceptions Across Gender Groups Using GPT-Based LLMs
- Oct 12, 2026Spectral-Driven Adaptive-Rank Optimizer for Large Language Model Reinforcement Learning with Verifiable Rewards
- Oct 12, 2026Large Language Models for Machine Translation Quality Annotation: Humans and Models Are Both Challenged
- Oct 11, 2026E8 Lattice Key Quantization for Low-Bit KV-Cache Compression
- Oct 10, 2026UltraBERT: Beyond the Limit of BERT Pre-Training
- Oct 10, 2026ProgAgent: Learning Tool Orchestration through Programmatic Training