SOURCE-LINKED INTELLIGENCE

TRACER: Per-Tool Context Retention for LLM Agents via Consequence-Attributed Reinforcement Learning

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Enterprise data agents answer business queries by chaining many tool calls over multiple reasoning steps, routinely accumulating hundreds of thousands of context tokens per session. Existing compression strategies typically allocate retention budgets without accounting for the downstream consequences of removing individual tool outputs. Aggressive compression may therefore trigger costly tool re-invocations that offset the initial savings. We call this the compression--consequence gap. To close it, we propose TRACER, which formulates compression as a sequential per-tool decision problem. A lig

Read original source ↗ Open in workspace

recordType

paper

region

Global