# When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors

Source: [When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors](<https://www.sota2.com/research/paper/69eaf8f355a3c32af89f8853>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-25T05:00:04.396Z

Authors: Chenghao Yang, Yuning Zhang, Zhoufutu Wen, Tao Gong, Jiaheng Liu, Qi Chu, Nenghai Yu

Published: 2026-04-23T03:48:56.000Z

## Abstract

Model distillation is a primary driver behind the rapid progress of LLM agents, yet it often leads to behavioral homogenization. Many emerging agents share nearly identical reasoning steps and failure modes, suggesting they may be distilled echoes of a few dominant teachers. Existing metrics, however, fail to distinguish mandatory behaviors required for task success from non-mandatory patterns that reflect a model's autonomous preferences. We propose two complementary metrics to isolate non-mandatory behavioral patterns: \\textbf{Response Pattern Similarity (RPS)} for verbal alignment and \\textbf{Action Graph Similarity (AGS)} for tool-use habits modeled as directed graphs. Evaluating 18 models from 8 providers on $\\tau$-Bench and $\\tau^2$-Bench against Claude Sonnet 4.5 (thinking), we find that within-family model pairs score 5.9 pp higher in AGS than cross-family pairs, and that Kimi-K2 (thinking) reaches 82.6\\% $S\_{\\text{node}}$ and 94.7\\% $S\_{\\text{dep}}$, exceeding Anthropic's own Opus 4.1. A controlled distillation experiment further confirms that AGS distinguishes teacher-specific convergence from general improvement. RPS and AGS capture distinct behavioral dimensions (Pearson $r$ = 0.491), providing complementary diagnostic signals for behavioral convergence in the agent ecosystem. Our code is available at https://github.com/Syuchin/AgentEcho.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Behavioral Similarity Analysis on τ-Bench and τ2-Bench (test)](<https://www.sota2.com/research/sota/behavioral-similarity-analysis-on-bench-and-2-bench-test.md>): Behavioral Similarity Analysis; τ-Bench and τ2-Bench (test); GED Score: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.21255>)

[PDF](<https://arxiv.org/pdf/2604.21255.pdf>)