--- title: "When Agents Look the Same — paper by Yang et al. (2026)" description: "The authors propose two metrics to fill this gap. 4 references. Among them: 0." authors: ["Chenghao Yang", "Yuning Zhang", "Zhoufutu Wen", "Tao Gong", "Jiaheng Liu", "Qi Chu", "Nenghai Yu"] year: 2026 venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)" source: "https://aclanthology.org/2026.acl-long.478/" doi: "10.18653/v1/2026.acl-long.478" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/yang-2026-when-agents-look-same-quantifying/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yang-2026-when-agents-look-same-quantifying/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/56/) / When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors # When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors **Authors:** [Chenghao Yang](https://md.research.elamur.ai/free/en/authors/chenghao-yang/), Yuning Zhang, Zhoufutu Wen, Tao Gong, [Jiaheng Liu](https://md.research.elamur.ai/free/en/authors/jiaheng-liu/), Qi Chu and [Nenghai Yu](https://md.research.elamur.ai/free/en/authors/nenghai-yu/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.478/) · [PDF](https://aclanthology.org/2026.acl-long.478.pdf) · [DOI 10.18653/v1/2026.acl-long.478](https://doi.org/10.18653/v1/2026.acl-long.478) · [OpenAlex](https://openalex.org/W7166823208) **Citations:** 0 citations (OpenAlex, 2026-10-14) ## Summary Many LLM agents now behave almost identically, likely because they are distilled from a few strong teachers. Existing similarity measures compare static responses. They cannot separate steps a task forces on every model from optional choices that reveal a model's own habits. The authors propose two metrics to fill this gap. Response Pattern Similarity (RPS) has an LLM tag trajectory segments with five interaction stages (authentication, elicitation, execution, verification, notification). An LLM judge then rates style, structure and alignment, but only on stages both models share. Action Graph Similarity (AGS) represents tool calls as directed graphs. It scores agreement on optional tools (S node), ordering habits (S seq) and output reuse (S dep). The authors measure 18 models from 8 providers against Claude Sonnet 4.5 (thinking) on τ-Bench and τ2-Bench. Same-family models are clearly closer to each other. Kimi-K2 (thinking) resembles Claude more than some Anthropic models do. A controlled LoRA distillation run confirms that AGS tracks the actual teacher. The framework offers a practical way to detect hidden behavioral convergence, which weakens the independence of agents. ## Key points - Same-family model pairs get AGS scores 5.9 pp higher than cross-family pairs. Anthropic models reach RPS above 3.8 with one another. - Kimi-K2 (thinking) is the most Claude-like non-Anthropic model: - AGS 82.7% and RPS 3.65. - S node 82.6% and S dep 94.7%, both above Claude Opus 4.1. - When both models solve the task, its S node with Claude is 0.661, versus 0.196 for GPT-5–Claude (3.4× higher). - Controlled distillation setup: Qwen2.5-14B-Instruct was fine-tuned via LoRA on 200 τ-bench trajectories from Claude Sonnet 4.5, with DeepSeek R1 as a non-teacher control. - AGS rose by +0.13 toward the teacher and fell by −0.05 toward the control. - The GED baseline rose toward both (+0.23 and +0.20), so it cannot tell them apart. - RPS and AGS are only moderately correlated (Pearson r = 0.491, p = 0.054, 16 non-Anthropic models), so they capture complementary signals. - Experiments use a generic ReAct framework. Code is released as AgentEcho on GitHub. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* ## Institutions - [University of Science and Technology of China](https://md.research.elamur.ai/free/en/universities/university-of-science-and-technology-of-china/) — University, Hefei, China · 145 works · 114,409 citations - [Nanjing University](https://md.research.elamur.ai/free/en/universities/nanjing-university/) — University, Nanjing, China · 99 works · 8,512 citations Also on the byline: [University of Science and Technology Chittagong](https://ror.org/00w9tx359). 📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works 🧭 Topics (1): [🪶 Efficient models & compression](https://md.research.elamur.ai/free/en/topics/efficient-models/) · 297 works ## Cite Cite as: Chenghao Yang, Yuning Zhang, Zhoufutu Wen, Tao Gong, Jiaheng Liu, Qi Chu and Nenghai Yu. 2026. When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{yang2026when, title = {{When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors}}, author = {Chenghao Yang, Yuning Zhang, Zhoufutu Wen, Tao Gong, Jiaheng Liu, Qi Chu and Nenghai Yu}, booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)}, year = {2026}, doi = {10.18653/v1/2026.acl-long.478}, url = {https://doi.org/10.18653/v1/2026.acl-long.478} } ``` - [← previous When 20 Agents Fail to Sort: The Distributed Sorting Benchmark for Scalable Multi-Agent Systems](https://md.research.elamur.ai/free/en/papers/yang-2026-when-20-agents-fail-sort/) - [next → When Bigger Isn't Better: Evaluating LLMs for Arabic Sentiment Analysis](https://md.research.elamur.ai/free/en/papers/ibrahim-2026-when-bigger-isn-t-better/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/56/) ## References 4 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Rishabh Agarwal, Nino Vieillard, Piotr Stanczyk, Sabela Ramos, Matthieu Geist, and Olivier Bachem. 2023. [GKD: generalized knowledge distillation for auto-regressive sequence models](https://md.research.elamur.ai/free/en/papers/agarwal-2023-policy-distillation-language-models-learning/). CoRR, abs/2306.13649. 2. Anthropic. 2025a. Claude-4.1-Opus: A large language model. [https://www.anthropic.com/](https://www.anthropic.com/) news/claude-opus-4-1. Accessed: January 2026. \#\# Acknowledgements Anthropic. 2025b. Claude-4.5-Sonnet: A large language model. https://www.anthropic.com/ news/claude-sonnet-4-5. Accessed: January 2026. [↗](https://www.anthropic.com/) 3. This work was supported by the National Natural Science Foundation of China (No. 62121002) and the advanced computing resources provided by the Supercomputing Center of the USTC. [find](https://scholar.google.com/scholar?q=This+work+was+supported+by+the+National+Natural+Science+Foundation+of+China+%28No%2E+62121002%29+and+the+advanced+computing+resources+provided+by+the+Supercomputing+Center+of+the+USTC%2E) 4. Victor Barres, Honghua Dong, Soham Ray, Xujie Si, and Karthik Narasimhan. 2025. [τ 2 -bench: Evaluating conversational agents in a dual-control environment](https://md.research.elamur.ai/free/en/papers/barres-2025-tau-2-bench-evaluating-conversational/). CoRR, abs/2506.07982. \#\# Limitations ByteDance. 2025. Doubao-1.6: A large language model. [https://www.volcengine.com/](https://www.volcengine.com/) docs/82379/1593702. Access… [↗](https://www.volcengine.com/) *Updated 2 November 2026*