2 papers
cs.CL2026
The Bitter Lesson of Tool Calling
Ishan Patel, Sahil Sen, Elias Lumer +1
Tool use transforms LLMs into agents that act beyond their training data, and for code-capable models, programmatic tool calling extends this further by replacing rigid JSON calls…
cs.LG2026
PolyKV: A Shared Asymmetrically-Compressed KV Cache Pool for Multi-Agent LLM Inference
Ishan Patel, Ishan Joshi
We present PolyKV, a system in which multiple concurrent inference agents share a single, asymmetrically compressed KV cache pool. Rather than allocating a separate KV cache per ag…