List your company

Tool selection on MetaTool similar choices subtask (test)

83.4AccuracyOATS-S1
41.07252.06163.0574.039Mar 13, 2026

Evaluation results

MethodLinks
OATS-S1Latency=3.5 ms, Hardware=CPU, type=emb. refine2026.0383.4
Vicuna-7bLatency=∼2–5 s, Hardware=GPU2026.0373.5
ChatGPT (GPT-3.5)Latency=∼1–3 s, Hardware=API / GPU2026.0369.1
Static Embedding (SE)Latency=3.7 ms, Hardware=CPU2026.0366.4
Vicuna-13bLatency=∼3–8 s, Hardware=GPU2026.0358.2
Average (9 LLMs)Hardware=GPU2026.0357
LLaMA2-13bLatency=∼2–5 s, Hardware=GPU2026.0344.1
BM25Latency=0.4 ms, Hardware=CPU2026.0342.7