# One Task, Many Interfaces: Simple Cross-Interface Consistency Training Improves LLM Agent Robustness

arXiv: https://arxiv.org/abs/2610.20576 · HTML: https://mlsanity.com/paper/2610.20576/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use.
- Builds on 31 of its 41 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Placed in

- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 30 of 31)

- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents](https://mlsanity.com/paper/2207.01206/) (2022)
- [Unsupervised Data Augmentation for Consistency Training](https://mlsanity.com/paper/1904.12848/) (2019)
- [τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://mlsanity.com/paper/2506.07982/) (2025)
- [Kimi K2: Open Agentic Intelligence](https://mlsanity.com/paper/2507.20534/) (2025)
- [GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models](https://mlsanity.com/paper/2410.05229/) (2024)
- [ToolRL: Reward is All Tool Learning Needs](https://mlsanity.com/paper/2504.13958/) (2025)
- [AgentTuning: Enabling Generalized Agent Abilities for LLMs](https://mlsanity.com/paper/2310.12823/) (2023)
- [ToolACE: Winning the Points of LLM Function Calling](https://mlsanity.com/paper/2409.00920/) (2024)
- [Grounding Large Language Models in Interactive Environments with Online Reinforcement Learning](https://mlsanity.com/paper/2302.02662/) (2023)
- [xLAM: A Family of Large Action Models to Empower AI Agent Systems](https://mlsanity.com/paper/2409.03215/) (2024)
- [Agent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models](https://mlsanity.com/paper/2403.12881/) (2024)
- [RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning](https://mlsanity.com/paper/2401.08326/) (2024)
- [TOOLVERIFIER: Generalization to New Tools via Self-Verification](https://mlsanity.com/paper/2402.14158/) (2024)
- [Procedural Environment Generation for Tool-Use Agents](https://mlsanity.com/paper/2506.11045/) (2025)
- [On the Robustness of Agentic Function Calling](https://mlsanity.com/paper/2504.00914/) (2025)
- [Tool Preferences in Agentic LLMs are Unreliable](https://mlsanity.com/paper/2505.18135/) (2025)
- [Generalizable End-to-End Tool-Use RL with Synthetic CodeGym](https://mlsanity.com/paper/2509.17325/) (2025)
- [Learning Evolving Tools for Large Language Models](https://mlsanity.com/paper/2410.06617/) (2024)
- [GenTool: Enhancing Tool Generalization in Language Models through Zero-to-One and Weak-to-Strong Simulation](https://mlsanity.com/paper/2502.18990/) (2025)
- [The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations](https://mlsanity.com/paper/2407.13957/) (2024)
- [What Do Agents Learn from Trajectory-SFT: Semantics or Interfaces?](https://mlsanity.com/paper/2602.01611/) (2026)
- [When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents](https://mlsanity.com/paper/2605.11928/) (2026)
- [Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environments](https://mlsanity.com/paper/2605.27209/) (2026)
- [Consistency Regularization for Domain Generalization with Logit Attribution Matching](https://mlsanity.com/paper/2305.07888/) (2023)
- [Can Agents Generalize to the Open World? Unveiling the Fragility of Static Training in Tool Use](https://mlsanity.com/paper/2607.01084/) (2026)
- [Do regularization methods for shortcut mitigation work as intended?](https://mlsanity.com/paper/2503.17015/) (2025)
- [Beyond Syntax: Action Semantics Learning for App Agents](https://mlsanity.com/paper/2506.17697/) (2025)
- [Do LLMs Know Tool Irrelevance? Demystifying Structural Alignment Bias in Tool Invocations](https://mlsanity.com/paper/2604.11322/) (2026)

## Abstract (arXiv)

Post-training binds the competence of an LLM agent to the interface it was trained on. Semantics-preserving perturbations of that interface, such as renaming an action or changing the notation, leave the underlying task unchanged yet make the agent worse at it. Scoring each sample independently means interface randomization only broadens this dependence from a single interface to the training interface distribution, so out-of-distribution perturbations still degrade the agent. We propose Cross-Interface Policy Regularization (CIPR), a self-supervised objective that replaces action names with uninformative identifiers and presents each training sample under two interfaces in which equivalent actions are identified by construction. CIPR penalizes the KL divergence between the action distributions induced under the two interfaces, so the policy behaves consistently across them. We evaluate CIPR in both supervised and reinforcement-learning post-training, where it improves robustness. On agentic function calling, held-out interface perturbations reduce the accuracy of a post-trained agent by nearly 25%, and CIPR recovers 10.5% while leaving accuracy on the unchanged interface intact.