[Submitted on 3 Sep 2026 (v1), last revised 22 Oct 2026 (this version, v3)]
Abstract:Tool-use evaluations typically measure calls through a serving stack, which can report zero calls even when model outputs contain call-shaped payloads. We study interface-induced trajectory censoring: a mismatch between model emissions and the interface that converts them into actions. On selected BFCL v4 subsets, changing only the serving adapter while holding model weights and other evaluation settings fixed moves scores under the official scorer from 0.00 to 0.96 on simple_python and from 0.00 to 0.19 on multi_turn_base. A template-parser factorial experiment shows strong complementarity: neither component replacement alone restores parsing from the baseline, whereas their joint replacement recovers 196/200 first-request parses. Across five Qwen2.5-Coder checkpoints, classifier-positive emissions increase with model size while the mismatched interface reports no calls; a separate matched-interface Qwen3 ladder exhibits few silent emissions through 8B. Experiments on 115 tau-bench retail tasks extend the mechanism to a stateful, non-code environment. In a preregistered, single-seed, 150-step 7B LoRA experiment, changing only the parser registered in verl increases tool executions from 10 to 16,844. However, neither a shared programmatic-feedback evaluation nor a post-hoc common native function-calling evaluation detects a held-out success gain from repair; rescue counts do not increase. Interface repair thus restores tool-mediated experience without a detected performance gain in this setting. These findings show that tool-use measurements depend on the model-interface configuration and motivate layer-wise observability and conformance checks before attributing evaluation or training outcomes to the model alone.
Submission history
From: Wenbo Wang [view email]
[v1] Thu, 3 Sep 2026 15:00:31 UTC (478 KB)
[v2] Wed, 21 Oct 2026 03:29:24 UTC (469 KB)
[v3] Thu, 22 Oct 2026 02:09:49 UTC (469 KB)
