Salvato in:
| Autori principali: | , , , , , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2602.07451 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908902527860736 |
|---|---|
| author | Zhen, Huiling Lin, Weizhe Liu, Renxi Han, Kai Li, Yiming Tian, Yuchuan Chen, Hanting Li, Xiaoguang Li, Xiaosong Chen, Chen Yu, Xianzhi Yuan, Mingxuan Yan, Youliang Qin, Peifeng Wang, Jun Wang, Yu Tao, Dacheng Wang, Yunhe |
| author_facet | Zhen, Huiling Lin, Weizhe Liu, Renxi Han, Kai Li, Yiming Tian, Yuchuan Chen, Hanting Li, Xiaoguang Li, Xiaosong Chen, Chen Yu, Xianzhi Yuan, Mingxuan Yan, Youliang Qin, Peifeng Wang, Jun Wang, Yu Tao, Dacheng Wang, Yunhe |
| contents | Diffusion large language models (DLLMs) have emerged as an alternative to autoregressive (AR) decoding with appealing efficiency and modeling properties, yet their implications for agentic multi-step decision making remain underexplored. We ask a concrete question: when the generation paradigm is changed but the agent framework and supervision are held fixed, do diffusion backbones induce systematically different planning and tool-use behaviors, and do these differences translate into end-to-end efficiency gains? We study this in a controlled setting by instantiating DLLM and AR backbones within the same agent workflow (DeepDiver) and performing matched agent-oriented fine-tuning on the same trajectory data, yielding diffusion-backed DLLM Agents and directly comparable AR agents. Across benchmarks and case studies, we find that, at comparable accuracy, DLLM Agents are on average over 30% faster end to end than AR agents, with some cases exceeding 8x speedup. Conditioned on correct task completion, DLLM Agents also require fewer interaction rounds and tool invocations, consistent with higher planner hit rates that converge earlier to a correct action path with less backtracking. We further identify two practical considerations for deploying diffusion backbones in tool-using agents. First, naive DLLM policies are more prone to structured tool-call failures, necessitating stronger tool-call-specific training to emit valid schemas and arguments. Second, for multi-turn inputs interleaving context and action spans, diffusion-style span corruption requires aligned attention masking to avoid spurious context-action information flow; without such alignment, performance degrades. Finally, we analyze attention dynamics across workflow stages and observe paradigm-specific coordination patterns, suggesting stronger global planning signals in diffusion-backed agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_07451 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DLLM Agent: See Farther, Run Faster Zhen, Huiling Lin, Weizhe Liu, Renxi Han, Kai Li, Yiming Tian, Yuchuan Chen, Hanting Li, Xiaoguang Li, Xiaosong Chen, Chen Yu, Xianzhi Yuan, Mingxuan Yan, Youliang Qin, Peifeng Wang, Jun Wang, Yu Tao, Dacheng Wang, Yunhe Computation and Language Diffusion large language models (DLLMs) have emerged as an alternative to autoregressive (AR) decoding with appealing efficiency and modeling properties, yet their implications for agentic multi-step decision making remain underexplored. We ask a concrete question: when the generation paradigm is changed but the agent framework and supervision are held fixed, do diffusion backbones induce systematically different planning and tool-use behaviors, and do these differences translate into end-to-end efficiency gains? We study this in a controlled setting by instantiating DLLM and AR backbones within the same agent workflow (DeepDiver) and performing matched agent-oriented fine-tuning on the same trajectory data, yielding diffusion-backed DLLM Agents and directly comparable AR agents. Across benchmarks and case studies, we find that, at comparable accuracy, DLLM Agents are on average over 30% faster end to end than AR agents, with some cases exceeding 8x speedup. Conditioned on correct task completion, DLLM Agents also require fewer interaction rounds and tool invocations, consistent with higher planner hit rates that converge earlier to a correct action path with less backtracking. We further identify two practical considerations for deploying diffusion backbones in tool-using agents. First, naive DLLM policies are more prone to structured tool-call failures, necessitating stronger tool-call-specific training to emit valid schemas and arguments. Second, for multi-turn inputs interleaving context and action spans, diffusion-style span corruption requires aligned attention masking to avoid spurious context-action information flow; without such alignment, performance degrades. Finally, we analyze attention dynamics across workflow stages and observe paradigm-specific coordination patterns, suggesting stronger global planning signals in diffusion-backed agents. |
| title | DLLM Agent: See Farther, Run Faster |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.07451 |