mlsanityData to 7 Oct 2026

Attributing Structured-Output Gains in Function Calling: Interface Alignment versus Procedural Transfer

arXiv 2607.02595 ↗First posted 1 Jul 2026cs.SE cs.PLcited by 0

On the map: LLM agents and tool use → Agent skill learning and evolution

Abstract (arXiv)

Structured-output benchmarks reward both task decisions and interface compliance, so prompt-induced function-calling gains require attribution before they can be interpreted as transferable skill. We introduce a four-layer gain-attribution protocol for prompt-prepended skill injection, combining canonicalized rescoring, format-only controls, repaired/balanced induction, and portability checks. Applied to the Berkeley Function Calling Leaderboard (BFCL) and scoped with API-Bank, MATH-500, and MultiHop-RAG, the protocol shows that several apparent gains are better attributed to interface alignment than to procedural transfer: format-only prompts match or exceed full skills in key BFCL cells, repaired/balanced induction removes the largest sub-frontier gains, and API-Bank target-native gains are matched within 0.5 percentage points (pp) by length-matched generic procedural prompts. These findings treat format compliance as a useful engineering capability while clarifying what a structured-output score certifies. We release BFCL-CANONICAL and recommend canonicalized metrics, balanced induction, and format-only baselines for function-calling skill-gain attribution. Code and data are available at https://github.com/couragec/skill-injection-attribution.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Agent skill learning and evolution, one of 16 directions in LLM agents and tool use. Builds on 13 of its 25 references that are in the map. First posted on arXiv 1 Jul 2026.

Builds on

13 of 25 references.

Closest ancestors · 1

Same direction, newest first

  1. 2026Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills26 Mar

Other references · 12

  1. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  2. 2023Let's Verify Step by Stepcited by 3,598
  3. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,880
  4. 2023Gorilla: Large Language Model Connected with Massive APIscited by 1,302
  5. 2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domainscited by 1,280
  6. 2023API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMscited by 574
  7. 2024MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queriescited by 253
  8. 2024Hammer: Robust Function-Calling for On-Device Language Models via Function Maskingcited by 53
  9. 2024NESTFUL: A Benchmark for Evaluating LLMs on Nested Sequences of API Callscited by 45
  10. 2026ContractSkill: Repairable Contract-Based Skills for Multimodal Web Agentscited by 13
  11. 2025Decoupling Task-Solving and Output Formatting in LLM Generationcited by 8
  12. 2026From Multi-Agent to Single-Agent: When Is Skill Distillation Beneficial?cited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026SkillAlign: Aligning Skill Interfaces for LLM-based Agents
  2. 2026Critical-State RL: Diagnosing Trainable States for Multi-Turn Tool Use
  3. 2026SAHOO: Safeguarded Alignment for High-Order Optimization Objectives in Recursive Self-Improvement
  4. 2025BLEUBERI: BLEU is a surprisingly effective reward for instruction following
  5. 2026Measuring Evaluation-Context Divergence in Open-Weight LLMs: A Paired-Prompt Protocol with Pilot Evidence of Alignment-Pipeline-Specific Heterogeneity
  6. 2026Brief Is Better: Non-Monotonic Chain-of-Thought Budget Effects in Function-Calling Language Agents
  7. 2025Inoculation Prompting: Instructing LLMs to misbehave at train-time improves test-time alignment
  8. 2026Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents