Robust Function Calling for LLM Agents Under Interface Drift

Tool errors or unavailable services cause your agent to abandon recoverable tasks.

What you would change

Inject repeatable tool failures and change only the recovery policy.

Inject repeatable tool failures and change only the recovery policy.

Compare against

Your current retry and fallback behavior under the same injected failures.

1. Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.
2. Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.
3. Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.
4. Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test.

What to measure

Recovered tasks, duplicate side effects, retries, time to recovery, and total cost.

When to keep it

Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation.

recommendation

test

research area

failure recovery

technical change

Validation-and-repair layer for function calling

published

2026-10-16

The method is training-free and lightweight, fitting constrained infrastructure, but its gains are demonstrated on two drift-focused benchmarks with a single 9B model, so the reader should validate it against their own tool inventory and drift patterns before production adoption.

1. claim 01Paperraft inference
   Read the brief interpretation

   SchemaShield-Lite replaces naive retry or model retraining as the response to interface drift in tool-augmented agents, wrapping the model's function-calling output with schema validation, candidate-set retry for missing tool names, and one-shot constrained repair. It costs additional inference calls on repair attempts, a modest increase in orchestration complexity, and depends on access to the candidate tool list and schemas at inference time. It can fail when the true failure is semantic rather than interface-level, when candidate-set drift is not the dominant damage mode in the reader's deployment, or when the repair call itself produces an invalid or wrong tool selection, although the paper reports no observed repair harms on its benchmarks.

   This is a Paperraft inference. It is not pinned to a verified PDF passage.

   Inspect the source
2. claim 02Paperraft inference

   On Qwen3.5-9B, raises DICE-300 accuracy from 0.717 to 0.813 and BFCL-200 accuracy from 0.580 to 0.805, recovering 19/22 and 33/45 drift-induced failures.

   This is a Paperraft inference. It is not pinned to a verified PDF passage.

   Inspect the source