Guardado en:
Detalles Bibliográficos
Autores principales: Li, Yanyang, Liang, Shuo, Lyu, Michael R., Wang, Liwei
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2408.03246
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916348477571072
author Li, Yanyang
Liang, Shuo
Lyu, Michael R.
Wang, Liwei
author_facet Li, Yanyang
Liang, Shuo
Lyu, Michael R.
Wang, Liwei
contents Recent advancements in long-context modeling have enhanced language models (LMs) for complex tasks across multiple NLP applications. Despite this progress, we find that these models struggle with multi-hop reasoning and exhibit decreased performance in the presence of noisy contexts. In this paper, we introduce Reasoning with Attributions, a novel approach that prompts LMs to supply attributions for each assertion during their reasoning. We validate our approach through experiments on three multi-hop datasets, employing both proprietary and open-source models, and demonstrate its efficacy and resilience. Furthermore, we explore methods to augment reasoning capabilities via fine-tuning and offer an attribution-annotated dataset and a specialized training strategy. Our fine-tuned model achieves competitive performance on multi-hop reasoning benchmarks, closely paralleling proprietary LMs such as ChatGPT and Claude-instant.
format Preprint
id arxiv_https___arxiv_org_abs_2408_03246
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Making Long-Context Language Models Better Multi-Hop Reasoners
Li, Yanyang
Liang, Shuo
Lyu, Michael R.
Wang, Liwei
Computation and Language
Recent advancements in long-context modeling have enhanced language models (LMs) for complex tasks across multiple NLP applications. Despite this progress, we find that these models struggle with multi-hop reasoning and exhibit decreased performance in the presence of noisy contexts. In this paper, we introduce Reasoning with Attributions, a novel approach that prompts LMs to supply attributions for each assertion during their reasoning. We validate our approach through experiments on three multi-hop datasets, employing both proprietary and open-source models, and demonstrate its efficacy and resilience. Furthermore, we explore methods to augment reasoning capabilities via fine-tuning and offer an attribution-annotated dataset and a specialized training strategy. Our fine-tuned model achieves competitive performance on multi-hop reasoning benchmarks, closely paralleling proprietary LMs such as ChatGPT and Claude-instant.
title Making Long-Context Language Models Better Multi-Hop Reasoners
topic Computation and Language
url https://arxiv.org/abs/2408.03246