A Study on Leveraging Search and Self-Feedback for Agent Reasoning
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_ | 1866909497770901504 |
|---|---|
| author | K, Karthikeyan Yuan, Michelle Mansimov, Elman Margatina, Katerina Pratik, Anurag Bonadiman, Daniele Sunkara, Monica Zhang, Yi Benajiba, Yassine |
| author_facet | K, Karthikeyan Yuan, Michelle Mansimov, Elman Margatina, Katerina Pratik, Anurag Bonadiman, Daniele Sunkara, Monica Zhang, Yi Benajiba, Yassine |
| contents | Recent works have demonstrated that incorporating search during inference can significantly improve reasoning capabilities of language agents. Some approaches may make use of the ground truth or rely on model's own generated feedback. The search algorithm uses this feedback to then produce values that will update its criterion for exploring and exploiting various reasoning paths. In this study, we investigate how search and model's self-feedback can be leveraged for reasoning tasks. First, we explore differences in ground-truth feedback and self-feedback during search for math reasoning. Second, we observe limitations in applying search techniques to more complex tasks like tool-calling and design domain-specific approaches to address these gaps. Our experiments reveal challenges related to generalization when solely relying on self-feedback during search. For search to work effectively, either access to the ground-truth is needed or feedback mechanisms need to be carefully designed for the specific task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_12094 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Study on Leveraging Search and Self-Feedback for Agent Reasoning K, Karthikeyan Yuan, Michelle Mansimov, Elman Margatina, Katerina Pratik, Anurag Bonadiman, Daniele Sunkara, Monica Zhang, Yi Benajiba, Yassine Artificial Intelligence Computation and Language Recent works have demonstrated that incorporating search during inference can significantly improve reasoning capabilities of language agents. Some approaches may make use of the ground truth or rely on model's own generated feedback. The search algorithm uses this feedback to then produce values that will update its criterion for exploring and exploiting various reasoning paths. In this study, we investigate how search and model's self-feedback can be leveraged for reasoning tasks. First, we explore differences in ground-truth feedback and self-feedback during search for math reasoning. Second, we observe limitations in applying search techniques to more complex tasks like tool-calling and design domain-specific approaches to address these gaps. Our experiments reveal challenges related to generalization when solely relying on self-feedback during search. For search to work effectively, either access to the ground-truth is needed or feedback mechanisms need to be carefully designed for the specific task. |
| title | A Study on Leveraging Search and Self-Feedback for Agent Reasoning |
| topic | Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2502.12094 |