Same Context, Different Agents: A Behavioral Study of Semantic Layers, Orchestration and Compaction in Data Agents
Answering questions over heterogeneous data means combining evidence from files, databases, documents and the web.
ProofPaper ↗
Key points
- Data agents automate this, but their success depends on context.
- Three context-engineering techniques (CETs) help by shaping it: semantic layers describe the data, orchestration splits work among agents, and compaction condenses interaction history.
- We present a comprehensive analysis of the three CETs, alone and combined, with seven large language models (LLMs) on 1,514 long-horizon tasks from five benchmarks, covering behavior, accuracy and cost.
- We find that data agents without CETs form three behavior groups: investigators, premature answerers that answer without inspecting data, and early stoppers that give up.
Sources (1)
- [1]Same Context, Different Agents: A Behavioral Study of Semantic Layers, Orchestration and Compaction in Data AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 14, 09:29 PM
Answering questions over heterogeneous data means combining evidence from files, databases, documents and the web.
Data agents automate this, but their success depends on context.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 14, 2026PN-Mix: Pair-Norm Guided Mixed-Precision Key Cache Quantization for Efficient Long-Output LLM Inference
- Oct 14, 2026RuleTrek: Benchmarking Agentic Rule Reasoning in Complex Scenarios
- Oct 14, 2026SafeDriveVLA: Navigation-Conditioned World Model Dreaming for Conflict-Aware End-to-End Autonomous Driving
- Oct 14, 2026Sparse Autoencoders Are a Shared Coordinate System for Symbolic Matching
- Oct 14, 2026Doc2Harness: Benchmarking LLMs on Long-Horizon Harness Engineering of Coding Agents
- Oct 13, 2026Plan or Retrieve First? A Scale Study of Tool Discovery Architectures
