Task-Oriented Key-Layer KV Communication for Efficient Latent Multi-Agent Collaboration
cs.LG
Submitted: 2026-09-24
Updated: 2026-09-24
Terminology
Sources
- Eliciting Latent Predictions from Transformers with the Tuned Lens
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Training Verifiers to Solve Math Word Problems
- Reducing Transformer Depth on Demand with Structured Dropout
- The Llama 3 Herd of Models
- Router-Tuning: A Simple and Effective Approach for Enabling Dynamic-Depth in Transformers
- DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
- DroidSpeak: KV Cache Sharing for Cross-LLM Communication and Multi-LLM Serving
- Adaptive Layer-skipping in Pre-trained LLMs
- Reasoning Capacity in Multi-Agent Systems: Limitations, Challenges and Human-Centered Solutions
- GPQA: A Graduate-Level Google-Proof Q&A Benchmark
- Fast Transformer Decoding: One Write-Head is All You Need
- Qwen3 Technical Report
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks