Completion $\neq$ Collaboration: Scaling Collaborative Effort with Agents

TL;DR

This paper advocates shifting from evaluating agents solely on task completion to assessing their ability to engage, collaborate, and enhance human effort over multiple interactions, emphasizing iterative, real-world problem-solving.

Contribution

It introduces the collaborative effort scaling framework to measure how an agent's utility increases with user involvement, highlighting the importance of sustained engagement.

Findings

State-of-the-art agents underperform in multi-turn scenarios

Current evaluations overlook collaborative engagement aspects

The framework helps diagnose and improve agent-human interactions

Abstract

Current evaluations of agents remain centered around one-shot task completion, failing to account for the inherently iterative and collaborative nature of many real-world problems, where human goals are often underspecified and evolve. We argue for a shift from building and assessing task completion agents to developing collaborative agents, assessed not only by the quality of their final outputs but by how well they engage with and enhance human effort throughout the problem-solving process. To support this shift, we introduce collaborative effort scaling, a framework that captures how an agent's utility grows with increasing user involvement. Through case studies and simulated evaluations, we show that state-of-the-art agents often underperform in multi-turn, real-world scenarios, revealing a missing ingredient in agent design: the ability to sustain engagement and scaffold user understanding. Collaborative effort scaling offers a lens for diagnosing agent behavior and guiding development toward more effective interactions.