TORSO: Template-Oriented Reasoning Towards General Tasks
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918141064380416 |
|---|---|
| author | Kim, Minhyuk Lee, Seungyoon Lim, Heuiseok |
| author_facet | Kim, Minhyuk Lee, Seungyoon Lim, Heuiseok |
| contents | The approaches that guide Large Language Models (LLMs) to emulate human reasoning during response generation have emerged as an effective method for enabling them to solve complex problems in a step-by-step manner, thereby achieving superior performance. However, most existing approaches using few-shot prompts to generate responses heavily depend on the provided examples, limiting the utilization of the model's inherent reasoning capabilities. Moreover, constructing task-specific few-shot prompts is often costly and may lead to inconsistencies across different tasks. In this work, we introduce Template-Oriented Reasoning (TORSO), which elicits the model to utilize internal reasoning abilities to generate proper responses across various tasks without the need for manually crafted few-shot examples. Our experimental results demonstrate that TORSO achieves strong performance on diverse LLMs benchmarks with reasonable rationales. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_09448 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TORSO: Template-Oriented Reasoning Towards General Tasks Kim, Minhyuk Lee, Seungyoon Lim, Heuiseok Artificial Intelligence The approaches that guide Large Language Models (LLMs) to emulate human reasoning during response generation have emerged as an effective method for enabling them to solve complex problems in a step-by-step manner, thereby achieving superior performance. However, most existing approaches using few-shot prompts to generate responses heavily depend on the provided examples, limiting the utilization of the model's inherent reasoning capabilities. Moreover, constructing task-specific few-shot prompts is often costly and may lead to inconsistencies across different tasks. In this work, we introduce Template-Oriented Reasoning (TORSO), which elicits the model to utilize internal reasoning abilities to generate proper responses across various tasks without the need for manually crafted few-shot examples. Our experimental results demonstrate that TORSO achieves strong performance on diverse LLMs benchmarks with reasonable rationales. |
| title | TORSO: Template-Oriented Reasoning Towards General Tasks |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2509.09448 |