Timely Machine: Awareness of Time Makes Test-Time Scaling Agentic
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915749853921280 |
|---|---|
| author | Ma, Yichuan Li, Linyang chen, Yongkang Li, Peiji Li, Xiaozhe Guo, Qipeng Lin, Dahua Chen, Kai |
| author_facet | Ma, Yichuan Li, Linyang chen, Yongkang Li, Peiji Li, Xiaozhe Guo, Qipeng Lin, Dahua Chen, Kai |
| contents | As large language models (LLMs) increasingly tackle complex reasoning tasks, test-time scaling has become critical for enhancing capabilities. However, in agentic scenarios with frequent tool calls, the traditional generation-length-based definition breaks down: tool latency decouples inference time from generation length. We propose Timely Machine, redefining test-time as wall-clock time, where models dynamically adjust strategies based on time budgets. We introduce Timely-Eval, a benchmark spanning high-frequency tool calls, low-frequency tool calls, and time-constrained reasoning. By varying tool latency, we find smaller models excel with fast feedback through more interactions, while larger models dominate high-latency settings via superior interaction quality. Moreover, existing models fail to adapt reasoning to time budgets. We propose Timely-RL to address this gap. After cold-start supervised fine-tuning, we use reinforcement learning to enhance temporal planning. Timely-RL improves time budget awareness and consistently boosts performance across Timely-Eval. We hope our work offers a new perspective on test-time scaling for the agentic era. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_16486 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Timely Machine: Awareness of Time Makes Test-Time Scaling Agentic Ma, Yichuan Li, Linyang chen, Yongkang Li, Peiji Li, Xiaozhe Guo, Qipeng Lin, Dahua Chen, Kai Computation and Language Artificial Intelligence As large language models (LLMs) increasingly tackle complex reasoning tasks, test-time scaling has become critical for enhancing capabilities. However, in agentic scenarios with frequent tool calls, the traditional generation-length-based definition breaks down: tool latency decouples inference time from generation length. We propose Timely Machine, redefining test-time as wall-clock time, where models dynamically adjust strategies based on time budgets. We introduce Timely-Eval, a benchmark spanning high-frequency tool calls, low-frequency tool calls, and time-constrained reasoning. By varying tool latency, we find smaller models excel with fast feedback through more interactions, while larger models dominate high-latency settings via superior interaction quality. Moreover, existing models fail to adapt reasoning to time budgets. We propose Timely-RL to address this gap. After cold-start supervised fine-tuning, we use reinforcement learning to enhance temporal planning. Timely-RL improves time budget awareness and consistently boosts performance across Timely-Eval. We hope our work offers a new perspective on test-time scaling for the agentic era. |
| title | Timely Machine: Awareness of Time Makes Test-Time Scaling Agentic |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2601.16486 |