# From Migration to Calibration: Aligning Agents with Deployment Requirements

arXiv: https://arxiv.org/abs/2609.35149 · HTML: https://mlsanity.com/paper/2609.35149/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 46 of its 50 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 30 of 46)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning](https://mlsanity.com/paper/1910.00177/) (2019)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning](https://mlsanity.com/paper/2205.12548/) (2022)
- [Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains](https://mlsanity.com/paper/2507.17746/) (2025)
- [ToolRL: Reward is All Tool Learning Needs](https://mlsanity.com/paper/2504.13958/) (2025)
- [ProcessBench: Identifying Process Errors in Mathematical Reasoning](https://mlsanity.com/paper/2412.06559/) (2024)
- [AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents](https://mlsanity.com/paper/2503.18666/) (2025)
- [SkillOpt: Executive Strategy for Self-Evolving Agent Skills](https://mlsanity.com/paper/2605.23904/) (2026)
- [Agent Lightning: Train ANY AI Agents with Reinforcement Learning](https://mlsanity.com/paper/2508.03680/) (2025)
- [In-the-Flow Agentic System Optimization for Effective Planning and Tool Use](https://mlsanity.com/paper/2510.05592/) (2025)
- [Distilling LLM Agent into Small Models with Retrieval and Code Tools](https://mlsanity.com/paper/2505.17612/) (2025)
- [Training Proactive and Personalized LLM Agents](https://mlsanity.com/paper/2511.02208/) (2025)
- [Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference](https://mlsanity.com/paper/2606.05922/) (2026)
- [Learning Personalized Agents from Human Feedback](https://mlsanity.com/paper/2602.16173/) (2026)
- [Learning to (Learn at Test Time)](https://mlsanity.com/paper/2310.13807/) (2023)
- [When Tools Fail: Benchmarking Dynamic Replanning and Anomaly Recovery in LLM Agents](https://mlsanity.com/paper/2606.05806/) (2026)
- [Reproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement Learning](https://mlsanity.com/paper/2606.04923/) (2026)
- [Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agents](https://mlsanity.com/paper/2608.15071/) (2026)
- [Step-wise Rubric Rewards for LLM Reasoning](https://mlsanity.com/paper/2605.17291/) (2026)
- [HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?](https://mlsanity.com/paper/2609.01437/) (2026)
- [HarnessOpt-Bench: Evaluating LLMs at Harness Optimization](https://mlsanity.com/paper/2608.06301/) (2026)
- [AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses](https://mlsanity.com/paper/2608.12307/) (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)
- [Reinforcement Learning for LLM-based Multi-Agent Systems through Orchestration Traces](https://mlsanity.com/paper/2605.02801/) (2026)
- [Harness-Zero: Harness Distillation via Agent-as-Harness](https://mlsanity.com/paper/2609.24974/) (2026)

## Abstract (arXiv)

Deploying, migrating, or scaling an agent changes its model, support system, application, and users. Calibration should align the agent with destination requirements without requiring source imitation or universal superiority. We propose three standard families: basic capability and operational reachability; technical and professional precision, including terminology, units, reporting periods, program fidelity, and report dependencies; and user context covering roles, use cases, authorized values, permissions, and budgets. A versioned mapping distinguishes requirements to preserve, transform, add, or retire. Four instruments---benchmarks, positive/negative examples, skills, and role-conditioned cost/utility---supply tunable reward channels within a fixed acceptance contract. Consultants annotate methods, branches, bounded loops, checkpoints, relative comparisons, and report refinement; a registry records applicable evidence. Adaptation revises configurations or trains a separate controller or skill while freezing the foundation executor. Independent hard gates govern qualification; RL and preference methods optimize candidates without supplying factual truth. Qualification and held-out improvement over the unchanged target are separate outcomes. Supplied intent labels permit a limited paired recount: small prompt/retry gains coexist with regressions and failed gates. FinQA supplements match prior summaries and retain provenance conflicts. These observations motivate destination-specific evaluation but do not establish calibration efficacy. We specify matched-budget tests, clean reload, complete-denominator reporting, and dependency revalidation. This is a methodological proposal with partially reconciled pilot evidence; confirmatory validation and the greater-than-80% zero-shot task-context acceptance target remain outstanding.